The best bond optimizers can create new dimensions for portfolio managers.
Technology is transforming bond investing, across research, trading and—through optimizers—portfolio construction. We believe optimizers based on advanced digital investment platforms have a major advantage—and can create new levels of insight for portfolio managers that have them.
Optimizers aim to create optimal allocations that ensure portfolios are aligned with their investment objective and incorporate their managers’ investment views, while also observing specified risk constraints and exclusions. They’re able to synthesize multiple investment dimensions simultaneously, including duration, yield-curve exposure, sector positioning, issuer concentrations and liquidity, plus many more factors.
But optimizers are only as good as their inputs. We believe that a leading-edge optimizer digitizes not only market data but differentiated research insights. This synthesis allows technology-empowered portfolio managers to iterate faster, refine portfolio outcomes more effectively and target specific portfolio exposures systematically.
That’s why it’s critical for an optimizer to be part of a suite of tech tools that work together seamlessly—collecting data, scoring bonds, assessing liquidity and providing two-sided pricing. Such integrated platform capability enables the optimizer to add value beyond routine functionality.
Taking Optimizers to the Next Level
Cutting-edge optimizers should also be able to process vast amounts of data nearly instantaneously. A bond issuer can have dozens of different bonds outstanding with varying maturities, coupons and covenants. An optimizer needs all this information, together with detailed bond-level analytics, including spreads, durations and risk sensitivities. The data set comprises thousands of bonds, each with a full set of analytics stored and refreshed daily. This isn’t something you can download from a website. Allocating resources to compiling, scrubbing and updating these data across global markets and over time is therefore a prerequisite.
But in our view, exceptional optimizers go beyond capturing clean and comprehensive market data by digitally incorporating the manager’s research insights—capturing bond-level and issuer-level scenario analyses. By embedding insights as direct inputs to an optimizer, investment managers can generate scalable, actionable buy and sell recommendations. But it doesn’t stop there.
A Fully Digital Process Enables Greater Versatility
Fully digitized advanced research processes can create many more outputs than just buy, sell or hold recommendations. And that wider range of outputs can translate into more effective optimizations.
For instance, a comprehensive set of quantitative and fundamental research findings that are condensed into single numerical scores from 1 through 100 (a “core score”) for each issue allows an optimizer to rank securities far more accurately than a score based on just a handful of options. And if that core score comes with upside and downside scenario risk ratings too, the optimizer has an additional dimension to work with.
Let’s assume an investment manager is building a credit portfolio with a high-yield objective. A standard optimizer would first focus on the highest-scoring high-yield bonds, then build in diversification and risk constraints that reflect the investment manager’s market views and risk tolerance. These constraints could include limits on interest-rate risk (duration) exposure, spread levels, and security and sector concentration levels.
However, we’ve found that an integrated digital platform can enable greater optimizer functionality and create more added value. For instance, by including upside and downside scenario risk data we’ve enabled our optimizer to tailor a portfolio for either a risk-on or risk-off market scenario, according to the portfolio manager’s expectations.
For example, a manager might observe that risks are rising in a late-cycle environment and instruct the optimizer to exclude any bonds to which their analysts have given a CCC downside rating. Our optimizer would then calibrate all these trade-offs and identify the optimal portfolio in seconds, rather than the days that manual iteration would involve.
Speed of Response Is Vital
Time is money in capital markets, and managers who are slow to respond to market events are often penalized. Portfolios that were optimal pre-COVID became sub-optimal once the pandemic struck, and likewise pre-trade-wars portfolios needed rapid reappraisal too. Now, the oil shock has created a third upheaval—all within a six-year time frame.
A digitized research process that can respond quickly and flexibly to such big-picture changes, and assess the resulting impacts on winners and losers, gives an investment manager a precious head start. And an optimizer that’s integrated with that process can capitalize on the advantage by recalculating optimal portfolios at once.
The Way Forward—Human and Optimizer Integration
Bond optimizers are a huge step forward—but that doesn’t mean an investment manager should simply delegate their portfolio construction to a machine. An optimizer built on a digitized research platform can create a super-rich data set that gives a portfolio manager many more ways to reflect their investment views comprehensively and precisely in portfolios.
Imagine a scenario where credit spreads have widened fast and the portfolio manager expects a reversal. A handful of bond managers now have tools to integrate that tactical view quickly, by capping portfolios’ exposures to spread momentum.
That’s not all. Our portfolio managers can observe these optimization changes and infer valuable insights over time about the interaction between different constraints and factors. They can also use the optimizer to create simulations and observe how portfolios constructed using different constraints would have performed.
For instance, energy is a big, volatile sector. An optimizer can give a manager the opportunity to test their human intuition regarding how tightly to constrain portfolios’ energy exposures against various simulated portfolios, across different scenarios and actual historical episodes.
It’s an amazing advance: we’ve found that our optimizer, working with our integrated digital platform, can create simulations that precisely reflect real-life conditions and can inform a portfolio manager how their decisions would have impacted performance under given circumstances. As a learning experience, that’s invaluable. And it means that an advanced optimizer can not only make portfolio managers faster, it can make them smarter too.