Heading into 2016, many Registered Investment Advisers (RIAs) continued a strategy in fixed income that had worked quite well since the financial crisis: overweighting credit and other higher-yielding sectors in search of higher returns. However, volatility in the first weeks of 2016 quickly drove high yield spreads almost 300 basis points (bps) higher, while U.S. 10-year Treasury bonds rallied by 70 bps.1 Reacting to this risk-off environment, investors collectively reallocated more than $100 billion to core bond funds and more than $20 billion to short-term and ultra-short bond funds over the first 10 months of 2016 ‒ representing more than 70% of the year’s net bond fund flows2 ‒ only to have the U.S. election results in November spur a Treasury bond sell-off that drove yields almost 80 bps higher. These events, coupled with PIMCO’s outlook for increased volatility over the cyclical horizon, are a reminder of the importance of understanding the key risks driving fixed income portfolio performance.
If fund flows alone tell the tale, many fixed income investors were
whipsawed by the events of 2016.
In our view, the sudden turnabouts are not over. PIMCO’s current outlook
for 2017 includes an increased likelihood of lower returns and higher
volatility. In addition to this difficult investment environment, many RIAs
are facing portfolio construction challenges that depend on understanding
and quantifying risk:
- Sources of risk in their portfolios
- Individual manager risks and how they have changed over time
- The impact of potential allocation changes on their portfolios
At PIMCO, we regularly conduct customized portfolio analyses for clients
using our proprietary tool MATADOR. Based on our review of 110 RIA
portfolios in 2016, we can provide insight into these three portfolio
construction challenges in light of the current investment environment and
considerations as advisers look ahead to 2017.
To start with an overall picture of fixed income portfolio risk, Figure 1
shows the average allocations by Morningstar category of PIMCO RIA clients
based on a portfolio analysis in 2015 and 2016. In 2016’s average
allocation, there was a noticeable increase in intermediate- and short-term
bond allocations compared with 2015 and corresponding decreases in
nontraditional bond exposures, in line with broader industry flows.
We used our proprietary analysis tool, MATADOR, to extract the 2016 average
portfolio risk factors, shown in Figure 2, to provide a much clearer
understanding of how the portfolio may react to changes in the markets.
Compared with the Bloomberg Barclays U.S. Aggregate Index, the average RIA
client portfolio analyzed by PIMCO is more diversified across sources of
risk, with a similar level of volatility. However, the average client RIA
portfolio analyzed is also more exposed to credit and emerging markets
(although these risks were marginally lower in 2016 than in 2015).
PIMCO’s MATADOR tool uses a rigorous analytical framework to help clients understand the key risk factors embedded in their portfolios. Through a sophisticated regression algorithm, the tool estimates intuitive risk factor exposures for funds and portfolios based on historical returns.
To find out how the average RIA client portfolio analyzed by PIMCO might
react to changes in yields and spreads, we used MATADOR to estimate the
portfolio’s average factor exposures – specifically, duration and credit
spread duration. We found that the average RIA portfolio analyzed in 2016
had a duration of 2.8 years whereas the Bloomberg Barclays U.S. Aggregate
Index had an estimated duration of 4.1 years. All else equal, for every 100
bps increase in the U.S. 10-year Treasury yield, the average RIA
portfolio’s value would decline by 2.8%. These factor exposures are a nice
complement to the portfolio risk decomposition and can help anticipate the
portfolio’s reaction to changes in the market.
Risks associated with individual managers
Identifying differences in risk across managers and how they may affect an
overall portfolio is a particularly useful analysis. While these
differences are often apparent in highly active funds, the comparison can
also be helpful in categories where investors are increasingly using
passive strategies. We reviewed two of these categories in detail,
intermediate-term core bond and short-term bond, using MATADOR.
First, we compared the risk and return profiles of the largest short-term
funds. As shown in the risk decomposition in Figure 3, the funds with the
lowest yields generally tend to have risk concentrated in short-term
duration. Most actively managed short-term funds add some high quality
credit to complement the passive interest rate exposure in short-term
government and agency securities and add potential return. While too much
credit risk can introduce unwanted volatility, especially during equity
market/risk asset sell-offs, the appropriate amount can help drive a
meaningful return difference in short-term allocations. Additionally,
short-term funds are most directly affected by increases in the Federal
Reserve’s policy rate, so allocating passively in these funds can result in
not only lower return potential but also limited flexibility to mitigate
the risks stemming from policy rate increases.
Similarly, in core bonds, the risk in the passive funds we analyzed was
largely concentrated in duration, whereas active funds often mix in other
types of risk, including credit, mortgage and currency risks, as shown in
Figure 4. These additional sector exposures can help active managers
navigate changes in interest rates and also potentially generate additional
return, which is particularly important today. For example, the yield of
the Bloomberg Barclays U.S. Aggregate Index has decreased to near all-time
lows, but its duration risk has increased approximately 20% since the
financial crisis. During the recent increase in bond yields from July
through December 2016, the Barclays U.S. Aggregate returned ‒3.3%, matched
by the average of the 10 largest passive managers who also returned ‒3.3%
while the 10 largest active managers outperformed by approximately 100 bps.
Impact of changing allocations within a portfolio
When allocating risk in portfolios, it is important to size allocations
appropriately in the current environment. For example, there are multiple
ways to add duration (at the short, intermediate or long end of the yield
curve), with varying effects on a portfolio’s overall risk profile. As
Figure 5 shows, the Bloomberg Barclays 1-3 Year Government/Credit,
Bloomberg Barclays U.S. Aggregate and Bloomberg Barclays Long Treasury
Indexes have significant differences in their total volatility, duration
and yield curve positioning.
Furthermore, the marginal impact of a change in risk allocation on the
portfolio can be significant – a 25% increase in short-term strategies, for
example, would have reduced total volatility in the average RIA portfolio
analyzed by 20% and placed more of the portfolio’s risk at the front end of
the yield curve, whereas adding 25% in long-term Treasuries nearly would
have doubled the portfolio’s volatility and concentrated almost all of the
portfolio’s risk in duration. As shown in Figure 6, these examples
illustrate the importance of quantifying the impact of potential portfolio
changes to ensure the desired effect on the portfolio’s risk profile.
While market volatility early in 2017 remains low, the events of 2016 and
PIMCO’s outlook for increased tail risks over the secular horizon make it
an important time to get a deeper understanding of your portfolio and
manager risks. This is often not an easy task, which is why we offer RIA
clients customized portfolio reviews using proprietary analytic tools like
MATADOR. Contact your PIMCO account manager for more information.
1 High yield represented by the Bloomberg Barclays U.S. High Yield Index.
2 Source: Morningstar