Indexology^® Blog

Most unconstrained bond funds claim to offer the following potential benefits:

• Low correlation to core fixed income;
• Attractive risk-adjusted returns; and
• Actively managed downside risk mitigation.

We examined each of these claims for the average performance of unconstrained bond funds since 2011 and noted that fund performance varied among them.

Persistently Higher Correlation to the Global Aggregate Bond Index Than the U.S. Aggregate Bond Index

Exhibit 1 shows the rolling two-year correlation of the average monthly return of unconstrained bond funds to that of the U.S. and global aggregate bond indices.  Though unconstrained bond funds do show periods of low, or at times negative, correlation to the U.S. Aggregate Bond Index, they also tend to demonstrate persistently high correlation of above 0.50 to the Global Aggregate Bond Index, though only until 2014.

Lower Risk-Adjusted Returns Than the U.S. Aggregate Bond Index on Average

Exhibit 2 shows the risk/return profile of unconstrained bond funds versus the U.S. and global aggregate bond indices.  On average, unconstrained bond funds delivered lower return and lower return per unit of volatility than the U.S. Aggregate Bond Index and higher return than the Global Aggregate Bond Index.  As average returns across funds tend to smooth out performance volatility due to the imperfect correlation between these funds, we also charted the performance statistics for quintile portfolios by return for the 36 funds that had full performance data for our analysis period.  Only one out of the five quintiles demonstrated a higher annualized return than the U.S. Aggregate Bond Index, and none outperformed the U.S. Aggregate Bond Index in terms of return per unit of volatility.

Improved Drawdown

Exhibit 3 shows the maximum drawdown for unconstrained bond funds versus the core fixed income indices for the analysis period.  On average, unconstrained bond funds experienced a maximum drawdown of 3.02%, which was better than the U.S. Aggregate Bond Index and the Global Aggregate Bond Index.  Exhibit 3 demonstrates the variance in maximum drawdown across all funds.  On average, at least 60% of funds experienced worse maximum drawdown than the U.S. Aggregate Bond Index.

According to the S&P Pan Asia Bond Index, India and China together represented 79% of the overall market value as of April 18, 2016.  While foreign investor access to these countries is opening up, the investability is still limited.  The market value excluding these two countries is around USD 2 trillion, as tracked by the S&P Pan Asia Ex China and India Bond Index, and Korea has the biggest exposure (see Exhibit 1).

Looking at historical performances, the S&P Pan Asia Bond Index outperformed the S&P Pan Asia Ex China and India Bond Index over the past one-, three-, and five-year periods, reflected the robust returns in the Indian and Chinese markets in recent years.  However, the trend has been reversing in 2016 and the S&P Pan Asia Ex China and India Bond Index increased 5.65% YTD, as of April 1, 2016. Interestingly, the S&P Pan Asia Ex China and India Bond Index also demonstrated higher volatility in the past 5 years (see Exhibit 2).

Exhibit 1: Country Breakdown of the S&P Pan Asia Ex China and India Bond Index

Some who follow target date fund (TDF) performance have taken note that lately, the S&P Target Date Index Series has outperformed many TDFs.  In most historical periods, index performance was middle of the pack.  However, 2015 was an exception, as shown in Report 2 of our Year-End 2015 Target Date Scorecard.  Every vintage of the S&P Target Date Index Series produced total returns that were close to, or better than, those of funds in the 90^th percentile of the target date mutual fund universe.

This story may seem like another chapter in the “index versus active” narrative, but it is particularly puzzling because of the methodology used to determine the benchmark asset allocation and glide path.  Every year, S&P DJI determines the asset allocation of each target date index vintage by compiling the holdings and asset-class exposure of a universe of active TDFs and using that data to create benchmark allocations reflecting average exposures of the active universe.  The benchmark therefore represents a consensus of asset-class exposure across its glide path; the average opinion, if you will, across the TDF industry.  I draw two conclusions from recent performance.

First, there is collective wisdom in the consensus.  There are a number of moving parts in every TDF suite.  Which asset classes are included?  What considerations inform the investment policy?  What assumptions are made?  How is the glide path expected to shift over time?  Which underlying funds are used to implement the investment policy?  What is the fee structure?  Whereas any individual TDF management team acts in response to an array of investment policy considerations, client interests, and self-interest, the consensus represents the wisdom of the crowd.  Collective thinking sometimes has superior value.

Second, it’s hard to beat a portfolio of index funds.  Single-asset-class managers and investors are challenged enough when it comes to beating cap-weighted benchmarks like the S&P 500^®.  The risk of active underperformance is increased when multiple funds are included in a portfolio.  Rick Ferri and Alex Benke demonstrate this effect in their paper, “A Case for Index Fund Portfolios.”

One final point about average performance: even if we never see another year like 2015, consistently achieving average performance is a more reliable way of compounding returns over time than alternately bouncing between outperformance and underperformance.  Perhaps tracking a market consensus glide path would be more enlightened than most industry insiders care to admit.

In 2015, the S&P BSE SENSEX remained volatile and ended in the red, at -3.68%.  In contrast, the S&P BSE India Government Bond Index ended the year in the black, at 8.33%.  The net investment by domestic mutual funds in the Indian equity and debt markets was significantly higher than the net investment by foreign investors.  Domestic mutual funds invested INR 721.98 billion and INR 4,340.99 billion in equity and debt markets, respectively.  Meanwhile, foreign investors added INR 178.08 billion and 458.57 billion, respectively.   Investors displayed mixed sentiments, let’s take a quick look at the performance of Indian large-cap equity funds from the latest SPIVA India Scorecard produced by S&P Dow Jones Indices and Asia Index Private Limited.

• Over the one-, three-, and five-year periods that ended in December 2015, 36%, 47%, and 57% of large-cap equity funds in India, respectively, underperformed the S&P BSE 100.
• Over the five-year period studied, the survivorship rate was low, at 70%, and the asset-weighted fund return was 103 bps higher than the equal-weighted fund return. This shows that the funds with a larger asset base had the advantage of the economies of scale over the five-year period.
• The return spread between the first and the third quartile break points of the fund performance was 4.59% over the five-year period, demonstrating that there was a wide dispersion in the returns of the funds in this category over this period.

Source: SPIVA India Year-End 2015 Scorecard.  Data as of Dec. 31, 2015.  Past performance is no guarantee of future results.  Table is provided for illustrative purposes.

For details on other periods and categories, please read the full report.

After China reported year-over-year first-quarter growth that showed signs of improvement, it overpowered negative news of the Doha oil production meeting failure and sent commodities rallying. Investors’ attention quickly shifted from oil to the other economically sensitive sector, industrial metals.

Many believe the uses for metals in construction and automobile manufacturing are the driving force behind the sector. Copper even earned its nickname “Dr. Copper” for its supposed ability to indicate the economic health of China. However, data shows Dr. Copper is not so smart and that this rally is from the weak dollar and supply shocks rather than Chinese GDP (gross domestic product) growth.

The first myth is that copper and industrial metals are correlated to economic growth, especially China’s. Using year-over-year data since 1978, the correlation of copper to Chinese GDP growth is only 0.21, and is actually the lowest of all the industrial metals. The most correlated metal is zinc but the relationship is still weak at only 0.28.

Even if Chinese demand might not be enough to boost the metals, there is still hope for the sector.  The correlation of metals to the US dollar has a higher (and inverse) relationship than to GDP growth, but the correlation is still only moderate.

However, the weakening dollar helps metals more than any other commodity sector. In the past 10 years, for every 1% move down in the US dollar, gains in lead, nickel, copper, zinc and aluminum have been 7.2%, 6.1%, 5.3%, 4.6% and 2.2%, respectively. Also, silver has gained 6.0% and gold has gained 3.5% for every 1% slip in the dollar.  The weak dollar helps far more than a strong one hurts.

Besides the tailwind of a weakening dollar, there are noticeable shortages that have appeared. March was the first month the industrial metals sector was in backwardation since September 2015 and marked the third time in 18 months – that hasn’t happened since 2007, before the financial crisis.  

Within the sector, there have been recent shortages in industrial metals from lead, copper and aluminum. While copper has more shortages than excess inventories throughout history, its roll yield has grown (measuring a shortage) 60% in the first quarter. Also, lead showed a shortage in Feb for the first time since Nov. 2012, but now is a seasonally weak time for lead as the winter demand for replacement automobile batteries slows. However, the support for aluminum could be more persistent from stockpiling and tax policies. It is very rare to see shortages in aluminum. There have only been 10 months in 10 years with a positive roll and it seems to be driving the whole sector. 

Notice the dramatic decline over the year in warehouse stock levels too.

Zinc has gained more than gold this year and is the best performer in the sector, up 20.5% in 2016. Its excess is half of what it was in October.  Supply cuts and mine closures have boosted returns. “Zinc inventories dropped for a 28th straight session to 413,250 metric tons, the lowest since August 2009, as supply cuts and the closing last year of China’s MMG Ltd. Century mine and Vedanta Resources Plc’s Lisheen mine boosted the need for supplies on exchanges. Chinese new-home prices rose last month in 62 of 70 cities tracked, compared with 47 cities in February, helping boost the demand outlook for the metal.” –  According to Bloomberg, stockpiles are the lowest in more than six years.

Nickel hasn’t seen a shortage since 2011, despite its huge price spike in 2014 of more than 50%. The suppliers are producing relentlessly to try to squeeze out marginal producers for market share – much like what is happening in the oil market. Except the role of China is flipped. Nickel producers want to squeeze China out, yet China is the oil customer everyone wants.

So, the story isn’t as simple as “Chinese demand growth boosts copper (or industrial metals).” While the demand growth may help, the dollar and supply side are the more important factors for the sector. If the demand grows at the same time the supply is disrupted and the dollar is weak, it may be a best case scenario for the industrial metals.