Canadian benchmark provides useful example of drawbacks and advantages compared to daily rebalancing
To closely reflect the market universe of securities they track, the benchmarks that underlie index-based funds undergo periodic rebalancing. But how often should that “periodic” rebalancing be, particularly among bond indexes?
That’s the question explored by a new research paper by FTSE Russell, which compared the daily-rebalanced FTSE Canada Universe Bond Index to a monthly-rebalanced equivalent. The two indexes follow the same criteria and constituent data, which covered the period from January 1, 2004 to December 31, 2019.
“The FTSE Canada Universe includes domestic bonds issued by Canadian corporations and Canadian federal, provincial and municipal governments,” the authors of the paper noted.
Focusing on the federal non-agency government bond sub-indexes for each benchmark, they found that average returns and information ratios were similar. However, a persistently lower duration in the monthly index over a period of mainly falling rates led to the increasing outperformance of the daily index over its monthly counterpart.
“Over the same period the annualized tracking error between the two indexes has been about eight basis points, based on monthly returns,” the researchers said. While the differences over time are not large, a gap in the timing of rebalancing between the two can result in appreciable changes in duration, particularly when a security with significant weight in the index is dropped or added.
“At such times, a sharp movement in yields could cause an appreciable difference in performance,” the researchers said.
Another consideration, they added, involves access to the primary market. Bonds are often cheaper to buy at issue than in the secondary market. Therefore, the choice to rebalance at a per-month pace can prove expensive as primary-market issues of bonds are missed.
Turning to corporate bonds, the authors noted that the FTSE Canada Universe Corporate Bond Index held just shy of 1,000 securities as of late 2019. Because of the much larger number of bonds, the indexes generated from daily and monthly rebalancing showed much smaller historical tracking errors.
“However, the persistent edge in average duration of the daily index evident in government bonds remains, albeit to a smaller extent, and once again produces higher returns,” they said. “With the secondary market in corporate bonds being markedly less liquid than that in government bonds, the attractions of the primary market are correspondingly greater.”
Monthly-rebalanced indexes, however, might be ideal for strategies that involve currency hedging. As the authors noted, most hedged indexes use one-month forward contracts to eliminate almost all currency risk, with the value of foreign currency exposure based on an assumed rate of growth in the index’s value.
“Hedging a daily index in the same way would not always be so effective,” the paper said. “New issuance and bond retirements would mean an index hedged in this way would be under or over-hedged most of the time, and therefore subject to an undue degree of currency risk and return.”