Lightnews — Scholar-powered news

Reposted by Nick Hagerty

Center for Effective Global Action @cega-uc.bsky.social · 1d

💧 Small-scale irrigation supports farmer livelihoods but can strain groundwater resources. How do we strike a balance?

Join us on Oct 23 for a session on Balancing Small-Scale Irrigation Solutions with Ram Fishman & @hagertynw.bsky.social.

Register: go.cega.org/registerATAIwebinar

1

Reposted by Nick Hagerty

AERE @aereorg.bsky.social · 14d

📢 Out now in the September 2025 issue of #JAERE! 📢
"Industrial Water Pollution and Agricultural Production in India" by Nick Hagerty ( @hagertynw.bsky.social ) and Anshuman Tiwari.
Read it here: buff.ly/oAalYIW
📈📉 #Econsky

3 6

Nick Hagerty @hagertynw.bsky.social · 15d

For people in Bozeman! We're voting on a ballot initiative this fall that sounds good but would cause some huge problems

(Precautionary disclaimer: we wrote this as private citizens, not representing our employer)

www.bozemandailychronicle.com/opinions/gue...

Kelsey Larson, Nick Hagerty & Justin Gallagher: WARD would hurt housing, environment

The WARD ballot initiative claims it will encourage sustainability and support affordable housing in Bozeman.

www.bozemandailychronicle.com

Nick Hagerty @hagertynw.bsky.social · 19d

Cool, maybe, thanks! I hadn't seen this paper and will have to take a look (and it didn't exist when we wrote the paper...)

1

Nick Hagerty @hagertynw.bsky.social · 19d

Last, industrial water pollution likely has many other costs to society. Our paper does not imply otherwise!

1

Nick Hagerty @hagertynw.bsky.social · 19d

Why doesn’t some of the worst industrial water pollution in the world seem to hurt agriculture?

Our best explanations:
- a lot of crops are less exposed than you’d think
- pollution gets diluted pretty quickly
- some pollution actually helps crops

1 1

Nick Hagerty @hagertynw.bsky.social · 19d

So: Here's how severely-polluting industrial clusters affect predicted crop yields. Downstream = positive distance = right side

You can see the damage, but it’s small (3%) and the conf. interval includes 0

Plus these are local effects that will decay further downstream

1

Nick Hagerty @hagertynw.bsky.social · 19d

To solve problem 3 (what areas are exposed) we develop new methods to carefully define and map upstream/downstream hydrological relationships

Defining these relationships is trickier than you might think. We discuss 4 ways prior methods can go very wrong

1

Nick Hagerty @hagertynw.bsky.social · 19d

This is hard because it’s the whole village, not individual plots with single crops

The result isn’t great as prediction (R2=0.25) but there’s meaningful signal as a proxy

AND our model is 4x better than just using NDVI, like many papers do

1 1

Nick Hagerty @hagertynw.bsky.social · 19d

To solve problem 2 (low-res ag data) we predict village-level crop yields from satellite data, training a random forest on nationally representative survey data.

1

Nick Hagerty @hagertynw.bsky.social · 19d

Industrial clusters coincide with a huge jump in river pollution of 300-600%. India’s government names and rates these clusters but doesn’t release public estimates of how polluting they are. We do

1

Nick Hagerty @hagertynw.bsky.social · 19d

To solve problem 1 (bad pollution data) we look at India’s giant industrial clusters and just estimate the effect of being downstream of a cluster. The spatial RD is obvious in Delhi here

1

Nick Hagerty @hagertynw.bsky.social · 19d

But 3 problems make this question hard to answer:
1. Pollution data is low quality
2. Ag data is low-resolution
3. Hard to know exactly what areas are exposed

1

Nick Hagerty @hagertynw.bsky.social · 19d

India has some of the worst industrial water pollution in the world. We wanted to know how it affects agriculture, because it’s often highly exposed and lots of people depend on it

1

Nick Hagerty @hagertynw.bsky.social · 19d

With Anshuman Tiwari

Link here: www.journals.uchicago.edu/doi/10.1086/...

Preprint here: hagertynw.github.io/webfiles/Wat...

Thread following

Industrial Water Pollution and Agricultural Production in India | Journal of the Association of Environmental and Resource Economists: Vol 12, No 5

Abstract We study how industrial water pollution affects agriculture in India, focusing on 48 industrial sites identified by the central government as “severely polluted.” We exploit the spatial disco...

www.journals.uchicago.edu

2

Nick Hagerty @hagertynw.bsky.social · 19d

Out today in JAERE! We measure water pollution released at India's industrial clusters. Does it hurt agriculture? Surprisingly: Not by much

Come for how we published a paper of null results & with no regression tables

Stay for new ways to proxy for crop yields & map hydrological relationships

🧵

2 9 16

Nick Hagerty @hagertynw.bsky.social · 27d

Do you think this is the relevant question? For whether I support radical tactics by elected officials on my side, it seems like what matters is whether I think elected officials on the other side are using radical tactics, not whether voters on the other side support them

Reposted by Nick Hagerty

Aaron Sojourner @aaronsojourner.org · Sep 8

🔥BLS Staff: "We will publish reliable data, no matter how inconvenient the results... Acting Commissioner is a respected career professional. There are no other political appointees at BLS... will remain accurate & nonpartisan... if that ever changes, the professionals will tell you." #StandWithBLS

“The Bureau of Labor Statistics must remain above politics. Our job is to deliver economic data guided by law and statistical practice — not partisan whim.

That’s why the firing of Commissioner Erika McEntarfer is so alarming. Dr. McEntarfer did nothing wrong. No serious observer claims otherwise. Commissioners don’t “cook” the numbers; they don’t even see them until after the estimates are complete. Dr. McEntarfer’s dismissal was unprecedented and she should be reinstated.

For decades, BLS Commissioners — Republican and Democrat alike — have been respected professionals. BLS data has been trusted because its methods are public, vetted, and transparent. BLS publishes its sources, publishes its methods, and its data revisions follow a set schedule. The public doesn’t have to *guess* whether the jobs numbers are real.

So why was Dr. McEntarfer forced out? Not because of misconduct. Not because of bad data. The only people making that claim are President Trump and his allies, who offer no evidence — only lies, character assassination, and intimidation. Their real goal is clear: to discredit independent statistics, slash budgets, and bully federal workers into silence.

But BLS staff will not be intimidated. We will publish reliable data, no matter how inconvenient the results. Our now-Acting Commissioner is a respected career professional. There are no other political appointees at BLS. The numbers will remain accurate and nonpartisan. And if that ever changes, the professionals will tell you.”

5 56 160

Nick Hagerty @hagertynw.bsky.social · Aug 26

AI can now do everything, which is just too tempting for even the most committed, intellectually curious students. I was probably as conscientious as they come and I certainly don't think I'd have been able to resist it

2

Nick Hagerty @hagertynw.bsky.social · Aug 26

I don't like it but HIGH FREQUENCY, IN-PERSON assessments are the only way college students are going to learn anything this semester

www.nytimes.com/2025/08/26/o...

Opinion | Students Hate Them. Universities Need Them. The Only Real Solution to the A.I. Cheating Crisis.

www.nytimes.com

1 2 10

Nick Hagerty @hagertynw.bsky.social · Aug 25

Link: jobs.montana.edu/postings/47791

Assistant Professor of Agricultural Economics

The Department of Agricultural Economics and Economics at Montana State University is hiring a tenure-track faculty member whose work focuses on the economics of agricultural production and farm manag...

jobs.montana.edu

Nick Hagerty @hagertynw.bsky.social · Aug 25

We are hiring at Montana State! Link in thread

If you aren't sure whether you fit the posted topic areas, please don't pre-screen yourself out, let our committee decide!

I'm not on the search committee but happy to answer any general questions

Demonstrated interest in scholarship relevant to commercial agriculture, applicable to commercial agriculture in Montana, especially in the following research areas: production economics, farm/ranch management, agricultural marketing, precision agriculture, energy economics, and financial risk management.

1 6 8

Nick Hagerty @hagertynw.bsky.social · Aug 17

Another approach might be spatially rolling averages? I totally buy that there's a lot of preference heterogeneity, I just expected it to vary more gradually across neighboring counties. But maybe you have reason to think counties are an important locus of sorting!

1 1

Nick Hagerty @hagertynw.bsky.social · Aug 17

Is that statistic for the mean coefficients you plot? I wonder if the year-specific coefficients are noisier - may make sense to shrink those original estimates before taking means

1 1

Nick Hagerty @hagertynw.bsky.social · Aug 17

Super cool! This seems like a perfect place for some empirical Bayes shrinkage. It's not really plausible that alternating counties in the Great Plains are extremely risk averse and risk seeking right? They're probably just noisy estimates which you could regularize to make a much nicer map

1 2