Our Privacy Statment & Cookie Policy

All LSEG websites use cookies to improve your online experience. They were placed on your computer when you launched this website. You can change your cookie settings through your browser.

April 30, 2020

Alternative Job Postings Data May Signal Analyst Downgrades

by Tim Gaumer.

Nearly every large company has long posted job openings on their own websites and those of third-party aggregators. We all know you can visit the sites to find out how many openings a company has and what roles they’re looking to fill. But, how would one know if the number of openings has recently increased or decreased? While tracing those changes may not be easy, we see value in doing so.

Companies often know more about their business conditions and customer demand long before external buy-side and sell-side analysts. Significant changes in anticipated demand are often met by preemptive responses like adjustments to hiring plans. The ability to observe this behavior could shed light on the direction of future revenue, earnings, and analyst revisions.

Take as an example, Applied Materials, or AMAT (its symbol). AMAT is a leading global manufacturer of semiconductor capital equipment (semi cap equipment). Its equipment is sold to and used by semiconductor (chip) manufacturers and foundries around the world.

It’s used for many steps in the process of turning a silicon wafer into memory and processor chips, such as depositing the material layers, plating and etching them in order to build the millions of diodes on each resulting chip. These fabulous and fabulously expensive machines are modern marvels whose creation requires the collective knowledge of material sciences, thermodynamics, and mechanical and electrical engineering. It’s also a notoriously cyclical industry.

Historically, when times were good, chip manufacturers all seemed to expand capacity at about the same time, ultimately leading to overcapacity. Overcapacity would lead to price competition and result in lower profit margins. Semi cap equipment companies would see order cancellations and join the downturn as there’s a strong correlation between chip makers’ revenue and profits and capex spending.

The current situation is different. It’s less a demand than a production, transportation and supply chain disruption. As Douglas Bettiner, CFO of Lam Research (one of AMAT’s industry peers) noted on the April 22, 2020 earnings conference call: “We are seeing the following dynamics. Capacity limitations are coming from our supply chain as well as adjustments in factory operations to maximize output considering social distancing challenges.” He added, “demand remains strong. We are output- constrained. These capacity challenges will negatively impact revenue and gross margins.”

Turning to another industry player, ASML Holding NV held its Q1 2020 earnings call on April 15. Chairman, President and CEO Peter Wennink responded to a question about the possible impact of COVID-19: “I think from a flexibility point of view, we have a lot of cost flexibility here with our variable cost on labor. It was a lot of variable cost, it’s like variable income to our people, which, of course, when the business goes down, that will go down also.”

Since Applied Materials’ quarter doesn’t end until the end of April, with reporting some weeks after that, we turn to alternative data to see if it could inform us about AMAT’s take on labor costs.

In Exhibit 1, we join data from third-party alternative data provider, Minneapolis-based LinkUp, whose job market data is now on our business partner’s BattleFin Ensemble Platform, with our own I/B/E/S Estimates data and Refinitiv’s quantitative StarMine data.

Exhibit 1: Applied Materials active job postings and revenue estimates, Nov. 15, 2019 to March 31, 2020

Sources: StarMine and I/B/E/S data from Refinitiv, LinkUp data, available on BattleFin Ensemble Platform

Job posting trends

The number of active job postings is data provided by LinkUp and shown in yellow. That number roughly doubled between mid-November to late February, growing from around 800 to a little more than 1,600.

The sell-side began raising revenue estimates around mid-January for the current quarter ending April 30, 2020. The chart displays two aggregate revenue estimates. The purple line shows the consensus, or I/B/E/S Mean Estimate. The orange line represents the StarMine SmartEstimate™ which is Refinitiv’s proprietary estimate that reweights the consensus, placing more weigh on more recent estimates and the more accurate analysts. Being predictive of the direction of future changes in consensus, we show it here as an early indicator of future analyst behavior. It worked as expected for AMAT. The orange line began a series of upward revisions before the consensus caught up about a month later.

Then in early to mid-March a number of active job postings for AMAT began to disappear, declining from around 1,600 to approximately 1,300 – roughly a 20% reduction. That would be our red flag.

It wasn’t until March 19 that we saw the first of the 20 analysts included in the I/B/E/S Mean lower his revenue estimate, citing the potential impact of COVID-19. On March 23, AMAT issued a press release, withdrawing previous 2020 guidance. By April 13, nine more analysts had lowered revenue expectations. Somewhat surprising, but not uncommon, the 10 remaining analysts have not updated the estimates that were published on February 12-13, shortly after AMAT last reported financial results.


While we don’t assume that every company will be quick to take down job postings as soon as hiring intentions change, Applied Materials appears to have. Visualizing that change, using LinkUp data, would have given investors a jump on even the first analyst estimate revision and actionable insights at least a week before management eliminated guidance and well before AMAT’s peers began providing commentary on current industry conditions.

Get In Touch


We have updated our Privacy Statement. Before you continue, please read our new Privacy Statement and familiarize yourself with the terms.x