Displacement Effects of Automation: Urban vs. Non-Urban Perspectives

By Aryaman PattnayakReviewed by Susha Cheriyedath, M.Sc.Dec 10 2023

An article recently published in the journal npj Urban Sustainability explores the displacement effects of automation technologies on urban versus non-urban labor markets. They argue that while automation displaces jobs across all settings, the impacts are unequal. Displacement primarily hits low-skilled workers in non-cities who may exit the workforce entirely. In cities, reinstatement and a shift towards higher-skilled, better-paid occupations offset adverse effects – potentially widening inequality.

Recent advances in AI, machine learning, robotics, and other technologies have raised longstanding fears about technological unemployment. Historically, automation has substituted for routine and manual tasks, depressing wages and job growth for less skilled workers. More advanced technologies now threaten even non-routine cognitive roles.  

Some predict that economic expansion may occur without employment gains. Economic and job growth have typically gone hand in hand, but automation enables production tasks to be handled cheaply and efficiently by machines rather than human workers. If technologies replace all job types in theory, overall economic productivity could increase even as human employment opportunities decline.  

Additionally, selective displacement and uneven impacts of automation across occupations are likely to amplify existing inequalities. As most tasks become automated, only a tiny fraction of workers stand to benefit significantly. For most, prospects will remain unchanged or decline, exacerbating divides between those automating and those automated. Technologies like AI systems may augment and complement high-skilled professional roles, creating new human-machine synergies. Simultaneously, automation eliminates swathes of jobs outright rather than generating new ones in their place.

This polarization looks set to depress labor's share of income. With technologies lowering firms' worker costs, returns concentrate in the hands of machine owners rather than human employees. Distributional impacts also loom as automation concentrates among lower-paid routine occupations. Consequently, an elite hyper-productive superstar workforce may reap automation's benefits while most laborers cope with job losses and eroding wages.

Despite significant research quantifying displacement effects from automation adoption – measured by industrial robot installation – the heterogeneity of impacts across geographic and settlement contexts remains underexplored.  

The Study

This paper examines how labor market consequences of automation technologies, specifically industrial robotization, differ across Italian urban and non-urban areas. Italy provides an exemplary case with varied settlement types – from metropolises to industrial and rural zones – and differential diffusion of automation technologies across macro-regions. Crucially, manufacturing retains a strong presence even in mid-sized and large cities.

The researchers utilize employment data from the Italian Labour Force Survey (RFL) 2009-2019, robot adoption statistics from the International Federation of Robotics (IFR), and a NUTS3 classification scheme to identify cities based on population thresholds. Their econometric models estimate robots' displacement effects on workforce participation, specifically on low, mid, and high-skilled occupational groups, while contrasting urban and non-urban areas. 

Italy's diversity of city types, automation adoption rates, and persistence of urban manufacturing make it apt to compare automation's consequences by geography. The research collects relevant occupational and robot adoption data at the national scale, leveraging the Statistical Office's standardized NUTS3 regional schema to parse urban from non-urban zones for analysis through regression techniques.

Findings

The analysis verifies that robots do substitute for human labor. Increased automation is associated with lowered workforce participation regardless of urbanization, countering notions that cities may be shielded. However, compositional impacts differ starkly between contexts.  

In non-cities, displacement primarily affects low-skilled employees, who exit from the labor market without offsetting reinstatement in similar roles. This aligns with automation viewing routinized manual occupations as prime targets for substitution. With few opportunities to shift into reinstated posts, displaced rural workers instead withdraw from participating.  

Despite comparable contractions in low-skill employment in cities, expanding high-skill job shares implies adverse effects concentrated amongst marginalized groups. Urban economies encompass abundant high-wage cognitive non-routine occupations, the upper-tier roles considered least susceptible to automation. Thus, while low-skill urban labor shrinks analogous to non-cities, growing elite job categories appear fueled by exiting vulnerable segments rather than upward transitions.  

Interpretations

The results suggest automation universally displaces workers, challenging assumptions that only manufacturing-intensive, non-urban areas are vulnerable. Italian cities' economic diversity and retained manufacturing roles insufficiently protect against substitution effects.  

Further research might examine impacts in areas linked to impacted cities or from automation extending to non-routine cognitive capabilities. If automation merely reshuffles tasks between regions and withdraws from manual roles balanced by big-city gains in skilled jobs, aggregate effects could be neutral. However, absolute employment is shrinking presently, and Italian data indicates urban winners and losers rather than direct transitions.

However, contrasting adjustments in occupational structure indicate that automation consequences remain unequal across space. Non-urban labor markets contract as displaced workers withdraw. Urban composition meanwhile shifts towards high-skilled, high-wage roles. Researchers posit that low-skill city dwellers become excluded rather than transitioning upwards as expanding elite job categories widen intra-urban inequality. 

Automation cannot be presumed to spare urban zones, acting as an inequality multiplier. While non-cities undergo outright workforce decline, cities restructure around privileged occupations. Workers transiting from automatable routine jobs to elite posts could theoretically dampen divides. However, researchers judge Italian evidence shows low-skill urbanites are pushed out entirely as upper-stratum opportunities widen for others.  

As automation permeates new ground like reasoning and creative roles, cities may need help adapting to workforces. Metropolises presently concentrate complex non-routine jobs, driving their inequality amplification. However, if cutting-edge algorithms and AI supersede knowledge-based professions, the privileged urban careers presently expanding could reverse course.

Future Outlook

Ongoing advances in automating technologies will further transform work compositions and skill demands. From self-driving vehicles to AI medical diagnosis software, automation continues to progress into non-routine tasks once considered capacities exclusive to humans. Researchers must clarify where new job creation occurs as the substitution scope expands.

These technologies may assume control of some sophisticated capabilities like information recall and synthesis. Nevertheless, machines lack ethics, creativity, empathy, leadership, negotiation, and strategic decision-making capacities. Such complex human strengths likely become increasingly valued by certain firms and roles. Targeted development of uniquely human skills could enhance career resilience.  

However, smooth workforce transitions cannot be assumed as automation evolves. Preparing balanced policy supports for displaced workers while carefully monitoring for concentration in high-end jobs remains prudent. If automation even modestly raises barriers to entering privileged echelons, benefits accruing narrowly to elite talent could further cement divides.

Ongoing tracking of automation's advancement across geography will reveal if introductions cluster in specific zones or permeate universally. Similarly, monitoring occupational impacts, especially tracing flows between low, middle, and high-skill tiers, will unpack distributional consequences. Comparing urban and non-urban settings has indicated an unequalizing direction presently. However, new equilibrating dynamics could still emerge.

Understanding precisely how these technologies progress, map across places, transform sector mixes, and reshape skill demands remains imperative for properly targeting supports. Ensuring automation proves empowering rather than excluding relies on research illuminating where automation hits, how work and workers adjust in response, and where any gaps in assistance emerge.