Golub Capital Social Impact Lab

YouTube video by SAReserveBank #G20SouthAfrica Side event — The implications of Al on the organisation of industry and work

Watch @Susan_Athey’s talk on the implications of AI on the organisation of industry and work at #G20SouthAfrica

www.youtube.com/watch?v=okDG...

“Governments will play a key role…in whether we actually develop the technology that will help lower-skilled workers become more productive by using AI to augment them with expertise that previously was difficult to acquire.”

The talk will cover:

✔️How AI is altering industry dynamics & structures

✔️How these shifts will impact public services such as health and education

✔️How AI market concentration could tax the global economy

✔️Why govt policy will be crucial in shaping AI competition and innovation

AI & digitisation are rapidly reshaping the way we work.

Policymakers need to understand how, and what to do about it.

Watch @Susan_Athey speak to G20 leaders about these issues tomorrow 16 July @ 13:30 CET. #G20SouthAfrica

bit.ly/3GyMFgm or bit.ly/44PTXFP

Beyond predictions, @keyonv.bsky.social also worked with @gsbsilab.bsky.social to show how these models can be used to make better estimations of important problems, such as the gender wage gap among men and women with the same career histories. Learn more here: bsky.app/profile/gsbs...

If we know someone’s career history, how well can we predict which jobs they’ll have next? Read our profile of @keyonv.bsky.social to learn how ML models can be used to predict workers’ career trajectories & better understand labor markets.

medium.com/@gsb_silab/k...

Estimating Wage Disparities Using Foundation Models The rise of foundation models marks a paradigm shift in machine learning: instead of training specialized models from scratch, foundation models are first trained on massive datasets before being adap...

Paper: arxiv.org/abs/2409.09894

Analyzing representations tells us where history explains the gap.

Ex: there are two kinds of managers: those who used to be engineers and those who didn’t. The first group gets paid more and has more males than the second.

Models that don’t use history omit this distinction.

We use these methods to estimate wage gaps adjusted for full job history, following the literature on gender wage gaps.

History explains a substantial fraction of the remaining wage gap when compared to simpler methods. But there’s still a lot that history can’t account for.

This result motivates new fine-tuning strategies.

We consider 3 strategies similar to methods from the causal estimation literature. E.g. optimize representations to predict the *difference* in male-female wages instead of individual wages.

All perform well on synthetic data.

Two extremes:

A representation that's just the identity function meets condition (1) trivially but not (2).

A representation that uses a very simple summary of history (e.g. # of years worked) should meet (2) but fails (1)

New result: Fast + consistent estimates are possible even if a representation drops info

Two main fine-tuning conditions:
1. Representation only drops info that isn't correlated w/ both wage & gender
2. Representation is simple enough that it’s easy to model wage & gender from it

Intuition: If working in job X at some point has a small effect on wages, but men are much likelier to have worked in job X than women, it may be omitted by a model optimized to predict wage.

Foundation models are usually fine-tuned to make predictions (like wages).

But representations fine-tuned this way can induce omitted variable bias: the gap adjusted for full history can be different from the gap adjusted for the representation of job history.

We use CAREER, a foundation model of job histories. It’s pretrained on resumes but its representations can be fine-tuned on the smaller datasets used to estimate wage gaps.

But this discards information that’s relevant to the wage gap.

In contrast, foundation models learn *representations*: lower-dimensional variables that summarize information.

Consider estimating the wage gap explained by differences in job history.

Job history is high-dimensional since there are many possible sequences of jobs. So most economic models describe histories using hand-selected summary stats (e.g., # of years worked).

Decompositions can inform policy: a large explained gender wage gap can suggest differences in choices or opportunities earlier in a worker’s career, while an unexplained gap may arise due to differences in factors such as skill, care responsibilities, or bargaining.

Ex.: estimating the gender wage gap between men & women with the same job histories.

A large literature decomposes wage gaps into two parts: the part “explained” by gender gaps in observed characteristics (e.g. education, experience), and the part that’s “unexplained.”

Foundation models make great predictions. How should we use them for estimation problems in social science?

New PNAS paper @susanathey.bsky.social & @keyonv.bsky.social & @Blei Lab:
Bad news: Good predictions ≠ good estimates.
Good news: Good estimates possible by fine-tuning models differently 🧵

The value of non-traditional credentials in the labor market This study investigates the labor market value of credentials obtained from Massive Open Online Courses (MOOCs) and shared on business networking platforms. We conducted a randomized experiment involv...

📄 Full paper (Athey & Palikot, 2024) on arXiv: arxiv.org/abs/2405.00247. We thank Coursera for their collaboration. Follow @gsbsilab.bsky.social for more research insights on the digital economy, education, and policy. #OnlineLearning #JobMarket

Takeaway 2: Our experiment isolates the effect for the least employable learners from credibly and systematically informing employers about online credentials. The positive finding helps build a case that such credentials may be good investments for workers seeking to transition jobs

Takeaway 1: Online learning platforms and professional networking sites (e.g., LinkedIn) can boost job outcomes with simple features. Even light nudges to encourage skill signaling (like sharing a certificate) can improve employment prospects

Who benefited most? The boost was greatest for learners with the lowest initial job prospects

Nor were the gains simply from sprucing up profiles. Treated learners’ LinkedIn pages weren’t more complete or active than the control group’s. This suggests it was the certificate signal itself – not a general profile update – that made the difference

We checked that this isn’t just a fluke of LinkedIn activity. The results held even after excluding any “new” jobs that started within 4 months of the intervention (to ensure we only count jobs found after the credential was shared)

And it paid off: nudged learners were ~6% more likely to land a new job within a year than the control group. They were also ~9% more likely to have a job in the same field as their certificate

To dive deeper, we looked closer at the subsample of ~40K learners who were LinkedIn users before the experiment. They were 17% more likely to add their certificate to LinkedIn if they received the treatment. Their profiles also saw more views, signaling higher interest from potential employers

After learners earned a certificate, we randomly assigned a subset to the treatment group who got a prompt to easily add their new credential to LinkedIn. The nudge worked. Treated learners added credentials to their LinkedIn accounts, and these certificates received visits from LinkedIn

We ran a global, randomized trial with ~800,000 Coursera learners who had earned certificates, and who either came from a developing country or had no college degree. Do they get more jobs if they link to their (verified) Coursera certificate on LinkedIn?