Summary: Many trade models fail to capture the full harm of tariffs. PWBM projects Trump’s tariffs (April 8, 2025) will reduce long-run GDP by about 6% and wages by 5%. A middle-income household faces a $22K lifetime loss. These losses are twice as large as a revenue-equivalent corporate tax increase from 21% to 36%, an otherwise highly distorting tax.
Revenue Impact: President Trump’s tariff plan (as of April 8, 2025) is projected to raise significant revenue—over $5.2 trillion over 10 years on a conventional basis (with micro-elastic responses) and $4.5 trillion on a dynamic basis (with economic effects). This revenue could be used to reduce federal debt, thereby encouraging private investment.
Comparison with a Corporate Tax Increase: Tariffs are estimated to raise about the same amount of revenue as increasing the corporate income tax from 21 to 36 percent, in the absence of these recent tariffs. While raising the corporate tax rate is generally seen as highly economically distorting, tariffs would reduce GDP and wages by more than twice as much. All future households are worse off. The estimated economic declines are likely lower bounds, with actual declines potentially even larger.
Broader Economic Impact: Many existing trade and macroeconomic models fail to capture the full harm caused by tariffs. Larger tariffs reduce the openness of the economy, including international capital flows. This is especially costly under the nation’s current baseline debt path, which is increasing faster than GDP, that is generally excluded from trade models or treated as neutral (Ricardian). U.S. households would need to purchase more bonds, requiring bond prices to fall (yields increase), domestic capital investment prices to fall (the marginal product of capital increases), or both. Even conservatively assuming only domestic capital investment prices fall, the reduction in economic activity is more than twice as large as a tax increase on capital returns that raises the same amount of revenue.
I notice this paper is from April 2025. Do you know if this group has done any updates in the intervening 6 months to show how well their model seems to be working? For example, are they able to determine yet in Table 2 what portion of the costs are being borne by consumers versus businesses?
Trump has sown chaos by altering tariffs on a whim, and that messes up the economy worse than predictably high tariffs. Businesses can function under high tariffs, but if tariffs change and there is constant uncertainty, low-margin businesses can make profits only by accident.
After the Twitter exit bump, the whole Mastodon ecosystem has dwindled to less than a million monthly users, distributed over hundreds or thousands of servers. Mastodon.social is filled with complete crap. It’s like enshittification without profits. Hacker News have more monthly users.
If you pick random active user, they often have zero active followers. They post their stuff into a void.
Every document parser must first answer this question:
Why not pandoc import filter.
--
The above does not mean: "do not parser outside pandoc". There are reasons not to use pandoc. It means: Please explain why not. It helps others to choose.
Maybe this is not the answer for Asciidocr, but quite often the answer is "pandoc has an internal representation of what a document is. That representation is not as rich as many of the formats it supports."
With 70%+ profit margins on data centers, it's enough that just $1.43 billion in data center revenue is secured for $1b investment to pay back.
Nvidia's larger investments are even more conditional. With OpenAI, the investment is contingent on building huge data centers in the future. If the bubble ends, Nvidia pays nothing; if it continues, Nvidia has locked in a customer.
No it’s almost certainly true. If you’d like some estimates you should read up on scott Alexander’s pre and post substack revenues. For writers with large followings substack is like a 10-100x monetization boost over other options.
Source: my publishing company published three books of Scotts essays during COVID, hit number one two and four on the essay category on Amazon, and generated less than one fiftieth of Scott’s first year of substack revenues based on his public comments.
Perhaps you missed the point of my story - publishing books is not lucrative. Or the second point - I have direct knowledge of economics for famous authors
All good as long as people don't think that modern industrialization means lots of jobs.
In all industrial economies, the number of people working in the industry declines, even if the industrial output increases. You could maybe increase the people working in industry by 1–5% for some time, but then it moves to a downward trend again.
Labor-intensive industries are low-productivity industries. Even China has started outsourcing them after Chinese wages have grown.
Computable systems can have have mathematically undecidable problems inside them.
Game of life is maybe the simplest example of simulated universe that contains many undecidable problems.
They fall into the same categorical mistake as the Lucas–Penrose argument, and they even use that argument in the paper. There is a lot of hand-waving. By the way, just adding irreducible randomness into a computational system would make it trivially non-computable in the meaning they use, but that itself would not prevent developing an axiomatic Theory of Everything that explains everything we want to know. So far, there has been nothing that demonstrates that the Universe must be non-computable.
Debt and over-leverage are the things that make the AI-hype problem for everyone. As long as AI investments are funded from revenue and wealth seeking profits, the bust makes a dent but takes very little from those who stayed away. With debt, you can create systemic risk.
Normal investment risk (risk of a stocks going down) is idiosyncratic, specific to that asset and can often be diversified away. Systemic risk is non-diversifiable because it is an external risk that drags down all interconnected parts, causing economic crises or financial contagion.
Massive debt (over-leverage) could create a systemic risk where the collapse of a few key firms forces banks to tighten credit, causing a wider recession that impacts all sectors. Systemic risk threatens the stability of the system itself, normal risk only threatens investors.
I did read the article, so here is one sentence debunk:
Computable systems can, and often do, have mathematically undecidable problems.
They fall into the same categorical mistake as the Lucas–Penrose argument, and they even use that argument in the paper. There is a lot of hand-waving. By the way, just adding irreducible randomness into a computational system would make it trivially non-computable in the meaning they use, but that itself would not prevent developing an axiomatic Theory of Everything that explains everything we want to know. So far, there has been nothing that demonstrates that the Universe must be non-computable.
Now Republicans are right radicals supporting state socialism. US government owns 10% of Intel.
reply