The level of threat to a given profession depends on two factors: the type of value provided and how it’s delivered.

There is no question that automation is changing the nature of work. But are the robots really coming for your job?

One of the most popular narratives is that low-paying jobs are doomed, while college-educated professions will remain largely untouched. Analysts often focus on wages and education as the primary predictors of job evolution, along with organizations’ potential to increase efficiency and reduce costs by changing or cutting jobs. But our research points to a more nuanced explanation.

A review of the academic literature and public discourse on automation revealed limited consideration of risks by profession. So we did our own comparison, coding 50 professions (including many from our literature survey) according to the type of value jobholders delivered and the skills they used to deliver it, to create a framework that helps workers assess what kind of threat automation poses for them. We identified four paths of evolution — jobs will be disrupted, displaced, deconstructed, or durable — and found that value is more predictive of change than wages, education, efficiency, cost, or other factors.

Counter to popular belief, it’s not necessarily blue-collar or non-college-educated workers who will be most threatened by automation in the coming decades. Our analysis suggests that a plumber may see less disruption than a legal professional. Simply instructing everyone to engage in continuous education and skill development is remiss. Workers must understand the four paths of job evolution — and the factors behind each path — if they hope to adapt.

Understanding the Four Paths

A jobholder uses a core set of skills to deliver value in some form to a recipient — either externally to a customer or within an organization. Jobs evolve as those consumers’ perceptions of value fluctuate along two dimensions: core skills and delivery mechanism, or what we call value form.

For some jobs, core skill sets include a specific knowledge base or craft. Others involve people skills and the ability to build relationships rather than technical expertise. Skills that can easily be standardized, codified, or routinized are most likely to be automated. Those that involve hands-on or real-time problem-solving are less so, because developing tools sophisticated enough to handle such ambiguity is either too cost- and labor-intensive or technologically out of reach.