PREVIOUSLY:

→ Guest article by Fred De Fossard of the Prosperity Institute: "Explained: The Afghan Migrant Scandal And Coverup Shaking Up Britain"

→ Analysis: How The Trump Administration Can Take Over The Ivy League

AND: don't forget to subscribe to the Sphere Podcast on Spotify, YouTube, or Apple Podcasts.

Trump Wins Everything On Trade

The biggest news today is of course the "trade deal" struck between President Trump and EU Commission President Ursula von der Leyen.

By universal agreement, it is a significant victory by the President, and the Europeans have essentially folded like a cheap suit.

In outline: under the agreement, the EU will face a 15% tariff on most imports to the US, including automobiles, pharmaceuticals, and semiconductors. Furthermore, the EU has committed to purchasing $750 billion worth of US energy products and invest an additional $600 billion in the United States, while also agreeing to buy substantial amounts of US military equipment. Here is Bloomberg's Lionel Laurent with an outline of the deal and the FT's Andy Bounds, Henry Foy and Ben Hall with a "tick-tock" of how the deal ended up being struck.

There are, of course, a lot of devils in the details. The non-tariff barriers, especially when it comes to agriculture, will be important to watch. It also doesn't seem to be physically or logistically possible for the EU to buy that much US energy, and it's unclear how the EU would "invest" $600 billion in the US given that it doesn't have a command economy.

It's also a missed opportunity for the US to make the EU embrace free speech.

But still, the overall picture is clear: after the UK, Japan, China, Vietnam, Canada, Mexico, and so many others, the EU, the world's biggest trading bloc, has caved to Trump.

Which is why it's worth taking a step back and seeing how much Trump was vindicated.

First, his negotiating strategy has been broadly successful. The American consumer really is the economic engine of the world, and other countries really are desperate to be able to tap into that engine, which means that fears of tariffs allowed him to negotiate much better terms than almost anybody else thought possible.

Secondly, there is a clear strategy here. Everybody screamed at the seeming incoherence of the "Liberation Day" tariffs, but they clearly worked as a negotiating gambit to get the US to the place where Trump wanted it, which is: 10% all-source tariffs, much higher tariffs on China, and higher tariffs or ad hoc measures to deal with specific industries or countries. That is what Trump wanted when he got into office, and that is where we are ending up.

Thirdly, the economic armageddon that has been universally predicted has failed to materialize. The economy remains stubbornly good, stubbornly posting numbers better than Wall Street and academic economic analysts predict, and prices remain stubbornly low. Store shelves are not empty. Supply chains have not collapsed.

Trump was right. He simply was right.

It's astonishing enough to spend time pondering.

Policy News You Need To Know

#Healthcare #TheScience #MAHA — "Evidence-based medicine." Who could be against that? However, the term EBM is a buzzword, and where there are nice-sounding buzzwords, dragons hide. Remember the old Ronald Reagan joke? "The trouble with our liberal friends is not that they're ignorant; it's just that they know so much that isn't so." Similarly, the problem with EBM is that so much of it isn't. The idea of EBM is that there is a hierarchy of evidence: some forms of evidence are more robust than others. At the top sits the queen of all medical studies, the randomized controlled trial with control group; at the bottom lies the poor old case study. And in between are all other forms of studies, especially observational and correlational studies. In theory, there's nothing wrong with this. RCTs are indeed the most robust form of studies, and case studies are just, well, case studies. The devil lies in the details, and in how the pyramid is used. The human body, and medicine, are so complex, that the edge cases are virtually infinite. Let's say a case study contradicts an observational study? Maybe the case study is wrong and the observational study is right. Or maybe the observational study was badly done, or just doesn't have much evidentiary power; meanwhile the case study has unearthed a special edge case that we should investigate more to understand. Here's another quote, this time from Jeff Bezos: "When the data and the anecdotes disagree, the anecdotes are usually right." In a world where most published research findings are false, the judgement, that is to say the anecdote-based pattern recognition, by an intelligent doctor with practical expertise with a certain pathology in a certain context may be a better guide to a successful remedy than a meta-analysis of a bunch of low-quality studies. Now, it's one thing for EBM to be one among other tools doctors can use, with its benefits and drawbacks like any other tool, or even a fad; it's another for EBM to acquire the virtues of gospel, which is what happens when EBM is used by the government or insurers to decide which treatments will be reimbursed and which won't. And this leads to a further problem. EBM, then, is easily manipulable: a bunch of studies near the top of the pyramid will beat what's in the below rungs of the pyramid, even if the studies are bad. And that creates perverse incentives once EBM can be used to move vast amounts of money. Anyway, this is why we welcome this article by Toby Rogers at the Brownstone Institute on EBM, explaining what it is, and how it was "hijacked by Big Pharma."

#AI — Sadly, the AI regulation moratorium failed to attach itself to the OBBBA. So now Adam Thierer is in City Journal talking about the threats of blue state regulation of AI.

#AI #Geopolitics — Speaking of AI, RAND—who, after all, developed the concept of Mutually Assured Destruction in the Cold War—have a paper looking at another form of game theory affecting global stability: whether the race to AGI could create conflicts.

#AI — Speaking of, again, R Street's Haiman Wong has a good paper out on the cyber and national security implications of the new AI Action Plan.

#MentalHealthCrisis — We talk a lot about the mental health crisis among teens. RAND has a new study out, published in JAMA, on how schools are tackling the mental health issue. "Researchers found that 30.5 percent of responding principals said their school required screening of students with mental health problems, with nearly 80 percent reporting that parents typically are notified if students screen positive for depression or anxiety. More than 70 percent of principals reported that their school offers in-person treatment for students who screen positive, while 53 percent of principals said they may refer a student to a community mental health care professional. The study found higher rates of mental health screenings in schools with 450 or more students and in districts with mostly racial and ethnic minority groups as the student populations." The study concludes that more should be done to screen kids for mental health conditions. To us, the finding that one third of schools systematically screen kids for mental health is an indicator that a lot of this crisis must be a statistical artifact.

#Nukes — New York Governor Kathy Hochul has restated her ambition to invest in nuclear to power New York state. Interesting times…

#Cybersecurity — Haiman Wong at R Street has an overview of the GUARD Act, which aims to tackle something which we feel will become an increasingly important problem, which is the use of the internet and AI to scam the elderly.

#Science #Technology — Neuralink, Elon Musk's brain chip startup, has operated on a new patient, who is now able to write her name on a computer with only her mind. This is simply an astonishing technological and scientific accomplishment, albeit one with concerning potential implications.