iiinova — where your digital self keeps evolving

asa watten @asawatten.bsky.social · 14/08/2026

If only there were 2 more centuries for the effects to be considerable!

asa watten @asawatten.bsky.social · 28/07/2026

I live in NOLA now and the sky does some stuff.

Lightning within a cloud in the distance. Tropical storm Bertha.

Sammy Roth @sammyroth.bsky.social · 23/07/2026

Reposted by asa watten

Insane new research. Scientists gave asthma patients free electric stoves to replace their gas stoves, then checked to see if symptoms improved. Not only did symptoms improve, ditching gas was *better than medicine.* Details in today's newsletter: www.climatecoloredgoggles.com/p/electric-s...

climatecoloredgoggles.com

Researchers used electric stoves to treat asthma. It worked.

The latest evidence that gas is bad for your health.

16175152808

asa watten @asawatten.bsky.social · 11/07/2026

though these scams demonstrate a real need for much better personal cooling tech

000

asa watten @asawatten.bsky.social · 11/07/2026

Great coverage of our working paper. 2 things I wish made it in the story: 1. Strong reasons to think that scaling up the grid reduces prices long run—excellent news for electrification. 2. The big risk with data centers is uncertainty—what if they don’t stick around to pay for all the new stuff.

152

asa watten @asawatten.bsky.social · 11/07/2026

… ope I dropped in on your commute.

John Bistline @bistline.bsky.social · 23/06/2026

Reposted by asa watten

To estimate whether data centers raise electricity bills, you need to solve a classic econometric problem: data centers pick low-price locations, which makes them look good. Our solution involves Dwight Eisenhower. New working paper with @asawatten.bsky.social and Geoff Blanford 🧵

183

Thanks for your interest! You can find more here: bsky.app/profile/asaw... And a summary here: restservice.epri.com/publicattach...

restservice.epri.com

000

We will write more about the details and caveats next week. If you can't wait until then, you can go read the paper (arxiv.org/pdf/2606.19777). There is also a summary here: restservice.epri.com/publicattach.... And we look forward to your (constructive) comments!

arxiv.org

If you're not in love with our instrument (that's OK!) the estimate is also negative (and small) once you control trends ("year fixed effects"). When you don't control for trends, they are positively correlated (not shown here). Elec prices went up, DCs too, and so did the width of jeans.

120

Our main results: - A doubling of DC capacity caused average prices to fall by 3.5% (2015 to 2024). - Using fewer, but more recent years, the estimate is 1.7%.

Why does this work? Fiber on land often gets built along highways. Access to fiberoptic trunks reduces operation and capital costs. And it turns out that they do a decent job of predicting DCs.

To get causal estimates we need something that drives data center investments that is unrelated to electricity prices. Our one weird trick: we use the 1947 Interstate Highway plan to predict where datacenters will locate > a half-century later. (We condition on state GDP and population).

So far this is not causal. The problem is that if you’re going to build a big, energy hungry thing, then you might consider the price of electricity when you choose where to build. This would lead to bias downwards. We could also tell stories that go in the opposite direction.

The incremental cost of generating new electricity is less than it used to be and certainly less than the average cost of the system. This can drive prices down. *The cost of wires has gone up. In CA they've gone up a lot due to wildfire.

This makes sense to us. We think the disconnect comes from the unusual way that fixed costs are paid for by customers in electricity markets. Fixed cost (wires, generators) are high and these costs are spread across customers and kwh. If these costs stay the same, more kwh means a lower ave cost.

... same is true if you look at percent changes.

We show: 1. DC capacity is not correlated with higher prices (2015-2024)

Figure of state-level data center capacity vs average residential price over time. States with the highest prices and price increases had lower than average data center capacity additions.

First I need to note a few things: 1. This is not peer reviewed yet. Science takes time. 2. Electrifying cars and heat is going to increase demand a lot.* We don't want this to increase prices. 3. Electricity prices are rising! So are DCs. But 2 trends does not imply a cause. *very diff load shape

asa watten @asawatten.bsky.social · 20/05/2026

🚨New preprint with @bistline.bsky.social and Geoff Blanford where we estimate the *causal* effect of data centers on avg residential electricity prices. DCs probably didn't increase your elc bill in the past. They certainly did not on average. Implications and caveats...🧵 arxiv.org/pdf/2606.19777

arxiv.org

131

And solar in the US employs roughly 6 Broadways or 80 coal industries

asa watten @asawatten.bsky.social · 15/05/2026

lots to be constern'd about with AI, but i can't help thinking much of the labor consternation stems from the engineering tendency to model problems as min cost given fixed output vs the econ tendency max output given fixed input for more nuance see: substack.com/home/post/p-...

substack.com

You are not a horse

AI and the future of labor demand

011

asa watten @asawatten.bsky.social · 13/05/2026

Of course there are all sorts of games afoot also...

000

asa watten @asawatten.bsky.social · 13/05/2026

It is shocking (intended) to think what's going on here is that the world has other options besides oil, unlike in 1973. And that's why the price shock is 1/3 while the supply shock is 3x. Looking forward to your papers in 2029 :p

132

asa watten @asawatten.bsky.social · 08/05/2026

Average residential electricity rates compared to inflation. While prices for goods are surely downstream of energy costs, it is notable that inflation spiked first and rates followed. Gray are individual states, black is US average, blue is CPI.

asa watten @asawatten.bsky.social · 07/05/2026

A decade of average residential retail prices.

asa watten @asawatten.bsky.social · 07/05/2026

A decade of average monthly residential electricity prices ($/kWh). A figure that defies a simple story.

000

asa watten @asawatten.bsky.social · 29/04/2026

Truly fascinating. While BLM keeps wind permitting on ice, they’re giving a green light to geothermal on public lands. Is this the beginning of a new ideological split split? www.blm.gov/press-releas...

blm.gov

BLM takes steps to accelerate geothermal energy development | Bureau of Land Management

The Bureau of Land Management has approved a categorical exclusion to make it easier and faster to explore for geothermal energy on public lands.

asa watten @asawatten.bsky.social · 26/04/2026

…but likely not so here. (Sustainable Aviation Fuels in CA. )

000

asa watten @asawatten.bsky.social · 26/04/2026

This can be a good thing if benefits in the new place or from the new type are much higher e.g. states with Renewable Portfolio Standards mandate %renewable electricity Federal subsidies for renewables (when they exist) are a transfer from rate payers to tax payers. Cold be good.

asa watten @asawatten.bsky.social · 26/04/2026

Some version of this point has been made for at least 20 years. And I guess economists still need to make it! The biofuel/renewables hokey pokey.

Fact 1: both renewable diesel and SAF are $2-$5 per gallon more expensive to produce than their fossil counterparts. Therefore, they need substantial subsidies to be cost competitive, otherwise they will not be produced.

Fact 2: the federal renewable fuel standard (RFS) requires that a minimum quantity of biomass-based diesel fuels be used in the United States each year. SAF counts as biomass-based diesel in the RFS. If the industry produces more biomass-based diesel than required, then the RFS subsidies will disappear. It has no incentive to go above the mandate because of the first fact. Therefore, more SAF means less renewable diesel in aggregate.

281

asa watten @asawatten.bsky.social · 21/02/2026

Oh oh force a divorce of sales and service. Terrible incentives and the main reason dealers steer ppl away from EVs (low maintenance). And bust up anti competitive service lock in. Just the worst.

John Bistline @bistline.bsky.social · 11/02/2026

Reposted by asa watten

Final EIA-860 is out, and 2025 really did it: 54 GW of new U.S. electric capacity, 96% clean. Solar again carries the offense, and we built more storage in 2025 than the cumulative total through 2023. More and more and more additions... but also record low retirements.

9333105

asa watten @asawatten.bsky.social · 12/01/2026

The logical fallacy that uncertainty => 0 is needs a snappy name.

asa watten @asawatten.bsky.social · 12/01/2026

$7.4M per life may be wrong, but $0 is extra super wronger. Hard to square with Section 812 of the Clean Air Act which requires EPA to conduct scientifically reviewed CBA that includes "overall health, welfare, ecological, and economic benefits".

nytimes.com

E.P.A. to Stop Considering Lives Saved When Setting Rules on Air Pollution

111

Electrons themselves barely move! On the order of meters per week in a reasonable DC circuit and hardly wiggle in AC. Electric fields traveling at ~light speed, directed by, but not in, wires, deliver energy. Pretty weird.

Mild take: 'Electrons' is simply a popular metaphor and that's fine 🤓 Electricity is profoundly unintuitive for most of us. Like water flowing through pipes is helpful, but wrong.

130

I would not be surprised if the pm effects are > change in traffic volumes because moving diesel trucks and busses from full stop over and over is real bad.

100

asa watten @asawatten.bsky.social · 08/12/2025

You have me convinced my prior should have low prob of PM effects >2x the change in traffic. And the method for the CI is sus. But parallel trends with other cities is violated for the reasons you mention (w-fire smoke, elec. gen) and they imply (local wind direction). Sometimes you just can't DiD.

120

The way back machine is great, but maybe it’s time to make forks and wikis of these sites e.g. @audreyt.org

030

asa watten @asawatten.bsky.social · 04/12/2025

you could call this take “don’t think of a framed elephant“

000

asa watten @asawatten.bsky.social · 31/10/2025

Also a bummer that higher ethanol blends appear to increase local ozone pollution. www.journals.uchicago.edu/doi/abs/10.1...

journals.uchicago.edu

University of Chicago Press Journals: Cookie absent