Interesting research doesn’t always come from procurement. Here are three findings from other disciplines that caught my attention this week, and what they might mean for sourcing and supply-chain decisions.

The Value of Having Another Option

New NBER research on production networks examines how the ability to switch suppliers changes exposure to economic shocks. Measures of risk can overstate vulnerability when they leave out the option to substitute one supplier for another.

For sourcing, that raises a practical question: how much is the ability to move volume worth when we evaluate an award? A second supplier may not offer the lowest price today, but the option to use that supplier tomorrow has value.

My take: Flexibility has value even when we never use it.

Read the research: The Robust Properties of Economies with Endogenous Production Networks (NBER, October 2026).

---

The Disruption Before the Disruption

Researchers using vessel-tracking data found that effective shipping capacity deteriorated months before major port congestion became visible. During recent disruptions, utilization losses reached 20 to 40 percentage points.

We tend to track disruption through delays, backlogs and congestion. But those can be symptoms of a problem that began much earlier. The useful question is what we could measure before the backlog appears.

My take: By the time the disruption shows up on the dashboard, the system may have been struggling for months.

Read the research: Shipping to America (NBER, September 2026).

---

The Expertise Behind the Equipment

Research linking machinery purchases with supplier transactions and worker skills found that much of the additional demand for technical expertise showed up outside the purchasing firms—in equipment integrators and later in complementary services such as software.

It’s a useful reminder that buying technology and realizing its value are different things. When we evaluate equipment, are we also evaluating the supplier capabilities needed to install, integrate and support it?

My take: The purchase order captures the equipment. It doesn’t necessarily capture everything required to make it work.

Read the research: Capital-Skill Complementarity Beyond Firm Boundaries (NBER, October 2026).

Three different studies, but a common question: are we measuring what matters to the decision, or just what’s easiest to see?