Davide Luparello

Unveiling Plant‐Product Productivity via First‐Order Conditions: Robust Replication of Orr (2022) We assess the replicability of Orr's (2022) method for estimating within-plant productivity across product lines, which combines demand estimation with cost minimization. The original study uses inpu...

Link to the published paper (early view):
onlinelibrary.wiley.com/doi/epdf/10....

When we extend to 2010–2019 and 2000–2019, IV demand estimation becomes unstable (inadmissible demand systems). We then calibrate demand parameters to the 2000–2007 estimates and still recover robust versions of the main results.
#EconSky #Productivity

We sweep the downstream-share IV exclusion threshold from 0–100%. Demand estimation moves around (see above), but the main implications are stable: large efficiency gains from dropping the worst product, and lower TFPR predicts product exit.

We rebuild Orr’s ASI Indian machinery data (2000–2007) and replicate the main patterns. Then we stress-test IV construction: we vary the downstream-share exclusion threshold used to select “other-industry” input price shocks. First-stage strength moves a lot.

Big picture: Orr (2022) shows how to recover plant–product productivity even when the data only report plant-level inputs—by combining (i) demand estimation with (ii) cost-minimization first-order conditions to infer product-level input shares.

Our replication paper with Joonkyo Hong is now published in the Journal of Applied Econometrics: “Unveiling Plant-Product Productivity via First-Order Conditions: Robust Replication of Orr (2022).”
📎 🧵👇

First R&R!

📈📉

First citation!

📈📉

Want to do a funded, joint Oxford-CMA PhD on market power?

Howard Smith (Oxford) and the CMA Microeconomics Unit are looking for promising candidates interested in doing research on the topic of "Beyond markups: understanding changes to competition, concentration, and market power".

#EconSky

1/2

Just back from #SEA! Huge thanks to Jon Williams for putting together such great IO sessions. Loved the presentations and discussions and looking forward to present my JMP there next time.

Now, back to the woods and my favorite activity (hint: it’s not econ…).

#econsky

The running list of IO job market candidates for 2024-2025 is live!

Fill this out to add your info: forms.gle/upCQ4Ez7sBTP...

Running list here: shoshanavasserman.com/io-jmc/

3/ This isn't just an economic tale but a societal one. How does the balance of productivity and worker welfare play out in one of India's biggest industries? Stay tuned for more insights from my work! 📈🚙

2/ My findings reveal how companies in the sector leverage considerable market power to keep labor costs low, boosting profits by reducing wages for blue-collar workers and favoring contractor agreements. It’s a story of progress, but with serious costs for those on the ground.🤔

1/ In this work, I dive into the rapid, exponential growth of #India's automotive sector over two decades. This boom has come with its own dynamics, particularly how car-makers’ market power impacts workers, especially contractors.

🎉 Thrilled to share that I won the Liberal Arts' RGSO Dissertation Competition at Penn State (again)! A (super short) 🧵 on my project: "A Tale of Power and Progress: Productivity, Markups, and Markdowns in India's Automotive." 🚙
📈📉

International Industrial Organization Conference | Industrial Organization Society

IO economists, the Call for Papers for the IIOC is out. It is May 3-5 in Boston. There are special sessions to showcase the work of grad students. Submit by Jan 31.

www.indorgsociety.org/conference

A great 🧵projecting the next decade of IO research on market power and technological progress

On my way to Malpensa feeling grateful. Big thanks to Comp-net organisers for inviting me to share my work on productivity & misallocation. A shoutout to Unibocconi trade group for the deep dive into markups & productivities. So many valuable insights. New drafts coming soon! 📈📉

For anyone looking to import their follows, chrome.google.com/webstore/det... worked great.