Levi Henze (@HenzeTimon) header
Levi Henze (@HenzeTimon) profile picture

Climate Economist with a passion for many other things. Swimmer and classical music nerd (apparently).

Berlin, de
Joined November 2012
Photo by Levi Henze (@HenzeTimon)Photo by Levi Henze (@HenzeTimon)Photo by Levi Henze (@HenzeTimon)Photo by Levi Henze (@HenzeTimon)Photo by Levi Henze (@HenzeTimon)Photo by Levi Henze (@HenzeTimon)Photo by Levi Henze (@HenzeTimon)Photo by Levi Henze (@HenzeTimon)Photo by Levi Henze (@HenzeTimon)Photo by Levi Henze (@HenzeTimon)
Technical explanation: AI models have thousands of stylistic tics that, if you had millions of examples of AI writing, could be extracted. The odds that a human has that combined set of tics is ~zero. This isn't particularly complicated or surprising theoretically. 3/3
1
2
2
162
5,756
Everyone wants to tax AI now. Congress: tax tokens 🪙 Senate staff: tax every replaced job 🔧 Bill Gates: tax tokens and robots 🤖 Europe: tax digital revenue 💵 We asked the economist's question: tax AI for what, exactly? New paper with Francesco Capozza. 🔗 👇 🧵1/9
3
43
4
152
13,604
“I 👏🏼 Want 👏🏼 Us 👏🏼 To 👏🏼 Remain 👏🏼 A 👏🏼 Slave 👏🏼 To 👏🏼 America”
An👏EU👏Army👏Is👏Literally👏The👏Last👏Thing👏 We👏Need👏For👏EU👏Defence👏
7
5
2
110
10,192
Europe's biggest car market is going fully electric. In Germany, battery-electric cars have now out-registered petrol and diesel combined for three months running. In October, BEVs made up 37.0% of new cars (94,086 of 254,492). Petrol + diesel made up 24.5%.
45
195
23
760
38,809
Wer denkt, dass wir ohne Steuererhöhungen (oder gar mit Steuersenkungen) oder Erhöhung der Schuldenquote Verteidigungsausgaben von 3,5% des BIP unterhalten können, lebt im Lala Land. Wie Verteidigungsausgaben historisch finanziert wurden zeigt @Ch_Trebesch kielinstitut.de/fileadmin/Da…
3
10
1
44
2,242
Levi Henze retweeted
Reality update
Replying to @Afinetheorem
Yeah, it happens, just not reliably. In OpenAI internal data, 8 hour tasks (measured by human length) suceed about 40% of the time. Getting to higher success rates shortens the time horizons dramatically.
1
2
129
"BloombergNEF warned data center gas plants could raise U.S. power emissions 10%"
17
24
8
131
12,989
Noch eine Ergänzung zu den Steuerklassen: Wir wollen eigentlich zu einer Welt, in der die korrekte Steuer jeden Monat einbehalten wird und keine Steuererklärung nötig ist. (Wer macht die schon gerne?) Wie kommen wir da hin?
32
10
7
84
6,215
Replying to @Vorname0Ziffern
Ist mein Job... 😉 Wenn die Änderung zu Steuerklasse iv/iv mit automatisiertem Faktor kommt, müssen Sie wahrscheinlich gar keine Steuererklärung mehr abgeben (vorausgesetzt Sie sind unter Werbekostenpauschbetrag und haben keine anderen Einkünfte). Das wäre doch was, oder?
8
2
1
30
2,340
My AI agents found an unmapped Maya site in Mexico, using LiDAR data from a NASA dataset. 72 hours from a random thought on vacation to a reply from Mexico's national archaeology institute. What a time to be alive…
264
2,050
620
20,059
1,372,656
I will be spending next week at the University of Vienna, presenting new work on the political consequences of low-carbon investments. Feel free to reach out if you would like to meet for coffee, lunch, or anything else!
2
8
2
74
5,200
Levi Henze retweeted
Replying to @majamediaco
arxiv.org/pdf/1407.1031 funny you mention that, people have been thinking about this for forever!
3
3
84
51,391
Levi Henze retweeted
today I discovered a walking route to work that’s 2 minutes longer than the maps default, but takes me past a park, beautiful architecture, and a street full of cafés and life. such a huge upgrade to a journey I make every morning I want Alive Maps, a version of Google Maps that optimises for beauty, serendipity, and aliveness instead of speed
🤖 Made with AI
245
182
87
7,586
1,395,558
Levi Henze retweeted
Em 1969 os russos estudaram o comportamento de animais em ambiente de gravidade zero. Vejam o gato. Simplesmente o gato!
Readers added context they thought people might want to know
O vídeo é de experimentos soviéticos documentados no filme "Внимание! Невесомость!" de 1964 (experimentos anteriores a 1961), não de um estudo de 1969. youtube.com/watch?v=kgmoq6… youtube.com/watch?v=0b0PWj…
395
1,369
104
25,828
5,236,306
Levi Henze retweeted
#EM Stocks Valuation Discount Deepens Versus S&P 500
1
2
2
221
Levi Henze retweeted
This is an absolutely fantastic interview question.
Omoh 😲🙆🏼‍♂️
209
1,299
45
42,981
2,287,793
Levi Henze retweeted
Post-Covid-19 inflation was predominantly driven by unexpectedly strong demand forces, both in the United States and in the Euro Area, from @DomenicoGiannon and @gprimice nber.org/papers/w35980
11
67
14
280
40,589
Levi Henze retweeted
Very cool paper. Under the Coase theorem, if transaction costs are low and property rights are well defined, you can get a Pareto efficient solution to negative externalities through a deal/payment. This is the case regardless of how the property right is allocated. In the case of global emissions, the disputed rights are a 'right to pollute' (i.e. the beneficiary must pay to get fewer emissions) or a 'right to no pollution' (i.e. the polluter must pay to emit more). This is the tricky political economy problem, which I also raised in my Coasean Bargaining at Scale blog post last year! The paper looks at how different rights allocations affect different countries. Some takeaways: 1. Standard climate models tend to use "Negishi weights," which assume zero international cash transfers under the guise of neutrality. The authors show this "no-transfer" rule isn't neutral at all and leaves certain polluters worse off. 2. Once the well-defined property rights are agreed upon, bargaining to the efficient outcome makes every country better off - unlike standard climate-economic models, which assume zero transfers. 3. Under “right to pollute,” the efficient allocation features total annual transfers of $190 billion, whereas under “right to no pollution,” total transfers amount to $1.03 trillion, which actually maps quite nicely to the sums negotiated at COP29. 4. As the largest emitter, China would have to pay ~$850B under the 'right to no pollution' regime, but would otherwise receive ~$95B under a 'right to pollute' regime. 5. Counter-intuitively, the US and EU+ would both be net recipients of transfers under "right to no pollution" given their low emissions relative to the size of their economies. 6. Some countries who emit more would effectively be worse off under a 'right to no pollution' regime (relative to doing nothing): South Africa (-7%), Russia (-4.5%), Eurasia (-4%), China (-4%), and Canada (-0.5%). 7. While switching property rights can be very significant for some regions (e.g. Sub-Saharan Africa would have to either pay ~$15B/year or actually receive $175B/year), ultimately there would be little impact on global inequality. nber.org/papers/w35971
4
15
1
103
8,013
My quick-and-dirty primer on the toolkit economists use for thinking about normative trade-offs, ie welfare economics. Seems relevant again …
Replying to @edenhofer_jacob
awareness of both its foundations and limitations. As for foundations: Welfare econ is committed to (i) consequentialism (utilitarianism is one type of consequentialist theory, but not the only one) and (ii) welfarism (👇). ➡️Consequentialism rules out the existence of
1
1
4
750
Replying to @brian_callaci
Economists should play a role in setting policy priorities above and beyond the practical implementation of a populist agenda, deferring economic goals to a simple majority is an abdication of meaningful responsibility
1
1
15
390