A day in the life of a bond index portfolio manager

Written by Karen Schenone, CFA
Fixed Income Product Strategist at BlackRock

Stratified sampling, platform risk, tracking error? Karen takes you into the world of bond ETF management.

Fixed income exchange-traded funds (ETFs) have continued to gather assets in 2018 as many investors are discovering their low cost and tax efficient benefits for the first time. But there are still a lot of misunderstandings out there, like this one: a bond index fund is a black box that robotically buys and sells bonds at the mercy of active investors. Nothing could be further from the truth.

Portfolio managers at iShares are bond market specialists, managing ETFs with the goal of tracking a bond index. And that’s not as easy as it sounds. To go behind the scenes of the making of a bond ETF, I’m talking to James Mauro and Karen Uyehara. James manages all the iShares bond ETFs as well as the broader ETF portfolio management team. Karen is a multi-sector portfolio manager who works on the day-to-day management for over 180 billion USD of assets (Source: BlackRock, as of 1/31/2018).

Schenone: What’s a typical day like managing a bond ETF? Walk us through what happens when you get into the office. What is your biggest concern about the ETFs?

Mauro: With growing trading volumes in bond ETFs, it is critical that we leverage technology to help scale our investment process. I focus on how to strengthen the trading process and risk controls to improve the bond ETFs we manage. That’s always top of my mind.

Uyehara: The first thing I think about in the morning is risk—risk in the portfolios and risk in the market. I look at Aladdin reports to check that all of my funds are in line with where I expect them to be. I also look at how the market is moving and think about how that may impact the management of the portfolios.

Schenone: How does technology help you make decisions about security selection?

Mauro: iShares bond ETFs are managed using a stratified sampling approach. Stratified sampling allows us to select a subset of bonds that have similar characteristics of a bond index, without having to hold every single bond. As bond ETFs become more popular, the more creation and redemptions of shares we have to manage. With minimizing tracking error and increasing operational efficiency in mind, we made a major investment in technology in the form of sampling algorithms (algos). Aladdin algos can quickly step through a decision-making framework based on defined investment parameters, helping us make faster and more accurate decisions on what securities to buy and sell to best match portfolio risk and returns.

Uyehara: The Aladdin suite of applications are very helpful in quickly evaluating the potential impact of proposed baskets of securities for creation and redemption activity. Aladdin tools allow an in-depth look at a portfolio down to the security level, and at the same time, give an overview of risk factors such as sector and subsector exposures, yield curve, ratings, etc. With this 360º look, we can model different scenarios in the portfolios and quickly see how they may be impacted from a risk standpoint, such as tracking error and the amount of cash.

Schenone: How do you decide to participate in a new issue versus sourcing a bond in the secondary market?

Mauro: At BlackRock our ETF managers are very cognizant of transaction costs associated with portfolio turnover. Trading in the secondary market incurs a transaction cost, which can be significant in corporate and high yield bonds. Participating in new issues is a way to source bonds without paying transaction costs.

Uyehara: We look at the pricing of the new issue and evaluate if the concession for new issue pricing is attractive. It depends on the size of the deals and the impact they will have on the benchmark. We are less inclined to invest in small deals with minimal impact to the benchmark and smaller concessions.

Schenone: How do you think about a creation or redemption of shares? What are biggest decisions that you have to make?

Mauro: Creations and redemptions are done primarily using an “in-kind” process where authorized participants (APs, or bond broker-dealers) deliver bonds in order to create shares or redeem bonds out of the ETF to redeem shares. In the example of an AP seeking to create shares in a bond ETF, they will propose a list of bonds for BlackRock to review. We will evaluate the list of bonds versus several screens to ensure index compatibility, then review the aggregate exposure versus the broad ETF across several risk factors. Our primary concern is that the “in-kind” bonds are a high-quality sample set of the existing ETF.

Uyehara: We want to make sure that any basket of securities we agree upon for creation or redemption purposes is a representative risk slice of the fund. While we make sure that the bonds selected are reflective of the overall risks of the fund/index, we also consider reducing any overweights or underweights that exist in the portfolio. For a variety of reasons, a fixed income portfolio won’t fully replicate its benchmark, but we use each creation or redemption basket to try to move the portfolio closer.

Schenone: Each month-end, many of the funds need to be rebalanced to adjust to changes in the underlying indexes. How much trading occurs? What can you do to mitigate portfolio turnover?

Mauro: Most bond indexes are rebalanced monthly. Rebalancing is a very important index event where new issues are bought in and exiting issues are sold out. Our portfolio management teams are organized by sector, where each sector specialist is keenly aware of rebalancing dynamics, turnover and trading costs that can impact their portfolios. Portfolio managers work with our Trading and Liquidity Strategy (TLS) team to optimize our participation in new issues and selling around month-end. The primary objective is to minimize market impact and costs to the fund.

Uyehara: Because of the stratified sampling approach, we don’t buy every issue held in the benchmark. Meaning, when we think about the risks of the benchmark, we put different risks into different buckets. The index might have 100 bonds that fall into a given bucket, but we could select just a few bonds to help meet the risk characteristics of that bucket for the portfolio. By matching the risk characteristics of each bucket, the portfolio as a whole should echo the benchmark’s key risks. We can avoid buying lots of small positions for the portfolio, which can help mitigate excess trading and transaction costs.