📗 Read our briefing here: ifs.org.uk/articles/wor...

📗 Read the journal article published in Labour Economics: ifs.org.uk/journals/imp...

12/ Curious how we estimate elasticities from job flows?

Or how correlated shocks make inequality worse?

Full story – theory, data, and projections – in our WP ⬇️

ifs.org.uk/publications...

11/ But this works both ways.

Occupations in health/education may see larger wage increases, but can’t attract IT workers – low cross-elasticity.

Result: IT projected to grow >20% as share of employment over the next decade.

10/ Take, for example, IT. Why does it grow fast?

Relative wage gains in IT attract inflows from jobs with smaller wage shifts.

It can pull in workers from many nearby jobs – both technical and managerial.

9/ The model projects

🏗️ IT & construction: employment gains
👩‍🏫 Health & education: wage rises
🏭 Manufacturing + some high-skill jobs (e.g accountants): wage declines

8/ We then simulate how the labour market might respond to automation-driven tech shocks. We combine

🧍 Worker flows (2012-2021)
🤖 Expert views on task-automation risk

= equilibrium projections of employment, wage & job flows.

7/ Key insight 2️⃣

Elasticity heterogeneity explains 20% of the variation in occupational employment growth over 1985-2010.
Demand shocks explain 60%, supply shocks the rest.

6/ Key insight 1️⃣

Demand shocks often hit groups of similar jobs (e.g. motor vehicle drivers, railway engine drivers). This makes it harder for workers to switch, slows employment adjustment, and leads to more unequal wage changes.

5/ We use German panel data tracking career moves.

[1985-2010] Some occupations adjusted to shocks via jobs (e.g. assistants), others via wages.

What explains this? Spoiler: elasticity matters.

4/ We develop a tractable equilibrium model of the labour market that accounts for this heterogeneity. It helps us:

1️⃣ explain historical employment & wage shifts
2️⃣ project responses to future automation shocks

3/ Labour supply elasticities differ not just by job but by job pair.

For example, workers in laboratory medicine may shift quickly into human medicine if wages rise – but carpenters into IT? Much harder.

2/ Not all jobs are equally “elastic”. Some (e.g. doctors) have rigid labour supply – higher wages don’t attract many new workers.

Others (e.g. assistants) are flexible – wage rises boost employment.