SYSTEMATIC ACTIVE EQUITY INSIGHT

New perspectives on macro investing

Sep 11, 2020

Techniques aiming to deliver excess returns have evolved significantly. Over the course of the last decade, BlackRock’s Systematic Active Equity (SAE) team has sought to capitalize on the changing landscape.

Much of the focus during this time has been on identifying differentiated alpha insights from non-obvious relationships uncovered via machine learning techniques in combination with novel sources of alternative data. While these techniques have become popular among quantitative equity investors, they are often beyond the reach of typical macro investors, those focused on the broad economic and geopolitical backdrop. These alternative tools only increase in importance during unusual times, such as the coronavirus crisis, allowing quicker analysis of a faster-changing investment backdrop.

Top-down and bottom-up

Because BlackRock SAE has already built the technology infrastructure needed to process massive amounts of data to produce investment insights for individual securities, we are able to use this approach to produce differentiated insights within the macro space as well. At its core, this approach is intuition grounded in economic fundamentals, supplemented by novel techniques and data sources, implemented in a systematic, unbiased fashion.

Using a rigorous research process, we seek to generate uncorrelated investment ideas that offer a strategic advantage in timing important sources of asset price returns that are impacted by broad economic trends. Ultimately, BlackRock SAE is able to marry security-level bottom-up insights with macro-level top-down views in order to inform our positions on major drivers of returns, such as country, industry and style factors.

We outline four important topics that are central to macro investing. Each of these helps form our differentiated approach to systematic macro investing.

1: Uncovering macro regimes

Quantitative investors typically look for strategies that have consistent performance over long periods of history, but don’t react quickly to changes in the macroeconomic environment. We employ dynamic strategy allocation that accounts for different market conditions, borrow techniques from the machine-learning community and literature to take a different approach for the identification of macroeconomic regimes.

2: Assessing monetary policy sentiment

To best capture sentiment around the monetary policy cycle, we have used natural language processing techniques on large corpuses of broker reports, news, publications and speeches by monetary policy committee members to attribute sentiment around future monetary policy decisions. The trickiest aspect of this exercise is filtering tens of thousands of documents down to those that are relevant to central bank behavior and specifically to individual countries’ policies.

3: Measuring macroeconomic uncertainty

Uncertainty about the macroeconomic environment is traditionally hard to measure simply because macroeconomic series are released at low frequency. At the same time, the behavior of companies and policymakers is deeply affected by macroeconomic uncertainty. SAE has developed innovative tools for natural language processing that have typically been applied to text and transcripts associated with specific companies. Measuring macroeconomic uncertainty is possible using similar techniques augmented with adjustments that allow us to isolate information pertinent to countries rather than companies and appropriately design text queries. We offer an example of our differentiated process at work.

Improving measures of macroeconomic uncertainty
Seeking to capture what the broad indexes miss

Improving measures of macroeconomic uncertainty

Source: Bloomberg and BlackRock SAE as of July 31, 2016. Past performance is no guarantee of current or future results.

4: Using alternative data to form differentiated macro views

SAE has broadened the search for alternative data sources, not only to pinpoint more granular and differentiated views on individual companies, but to improve our understanding of the macroeconomic environment. We are often able to capture more timely and granular views on macroeconomic conditions by using data that has a direct link to the broader economy, such as to track consumer sentiment, government contracts to reflect government spending, or maritime shipment counts to capture trade activity.

Download full report