Tariff time

Still mostly wrong

I live in New Zealand, where I work as a specialist physician. One of the many enjoyments of my job is listening to medical students. I learn a lot. The sixth year of their training, just before they get their degree, is an “Internship”. One such intern was perhaps the most quiet person I’ve ever met. Let’s call him ‘Mark’. Mark “Wouldn’t say Bo to a goose”.¹ But surely there’s more under the surface? Let’s see.

I like to take a break in the middle of my morning ward round, so one day I sat him down and over coffee asked a common question, “What are you going to do after graduation?” His answer surprised and intrigued me: “Well, I don’t think I’ll do a job as a house officer [The next logical step]. There are just too many opportunities out there”.

He was initially a bit reticent, but I guess I have an open and honest demeanour, so he soon revealed all. It turned out that during his internship, Mark had made a million dollars! He was quite comfortable explaining how he did this.

Mark had spotted something nobody else had. The starting point for his reasoning was “You can sell anything on eBay” (apart from body parts). He also noticed that at the time, people tended to hoard Internet domain names, waiting for ‘a good price’. So he decided to do something different.

Every morning before breakfast—and before proceeding to his not-too-well-remunerated internship—Mark would receive a free list of all newly-expired Internet domain names. He ran these through a little Python program he’d written, which would filter out the rubbish: Compu-Global-Hyper-Mega-Net.com is out. He’d then work through the rest by hand, and buy promising names in bulk for about $5 each (as I recall). On average, about fifty per day.

But rather than waiting for that ‘really good price’ to materialise, he’s sell them all on eBay for $60. For that price, pretty much every one would be rapidly snapped up. And 18,200 transactions later, he’d made a million dollars. I think that apart from my open and honest demeanour, the reason why he was happy to share this is that after a year, others were catching on, so he was sure to move onto something new. The first million bucks is always the most difficult, anyway.

Moving a couple of decades on, we now have Trumponomics. Which made me think of my medical student again. Specifically, let’s look at how Trump’ team calculated those tariffs.

The Republic of Markistan

By now we’ve all seen that formula:

\(\Delta\tau_i = \frac{x_i-m_i}{\epsilon\phi m_i}\)

Here’s the original brainiac post by Kush Desai (Screenshot taken from an X post by James Surowiecki) :

As we know, this ugly mess reduces in their hands to “Calculate the tariff as follows: divide the trade deficit by the imports”. In that equation, x is the exports and m the imports; they conveniently choose values for epsilon and phi that cancel out! Well, let’s crudely do the calculation for Mark’s little business. I’m going to pretend he’s a small country: the Republic of Markistan.

We’ll make some assumptions: that Markistan bought domain names from a US supplier, and sold most of them back to people in the US. Let’s also pretend that these are ‘goods’. Now, of course, this is simplistic. In reality, other small countries that have nothing to do with the US (Janeistan, Fredova) might have bought domains from Markistan for $60, but let’s assume the bulk of his commerce was with the US. And we can argue about other small details. But bear with me.

So Mark has bought ‘goods’ worth $91k, and the US imported them back for $1.092M. The difference between the two is almost exactly $1M, giving a Trumpian tariff rate of $1M/$1.092M, or about 92%. Trump then shows how kind and forgiving the USA is, dividing by two to produce a final tariff rate of 46%. Markistan should be grateful.

Trumpistan

Which naturally speaks volumes about the intellectual wherewithal of Trump’s team of sycophants, and their enviable ability to troll through ChatGPT for world-wrenching economic policies. And their ignorance that \times gives you a multiplication sign in LaTeX (and that the *s are not needed anyway). But this is small beer. There are bigger issues here.

First, can’t you argue that in a sense—especially with my illustration—that Trump’s team is right, here? All Markistan did was sell the Internet domain names back to “the US”? Isn’t this cheating? Doesn’t this small country deserve to be punished (or at least taxed) for this ‘skulduggery’? This perception illustrates the difference between reality and, well, Trumpistan.

At the end of the day, it’s not about the balance of trade. It’s about who has benefitted. And it’s obvious here that Markistan benefitted—but it’s equally obvious that everyone who obtained a useful domain name at a very reasonable price also benefitted. Mark took raw material, and through the simplest and yet most profound means, transmuted it into value for everyone. Win-win. And equally obviously, Trump doesn’t get this. In contrast, here’s what Trump said on the afternoon of Wednesday 2 April 2025—off by a day and a half:

For decades, our country has been looted, pillaged, raped, and plundered [sic] by nations near and far, both friend and foe alike. American steelworkers, auto workers, farmers, and skilled craftsmen. They really suffered gravely. They watched in anguish as foreign leaders have stolen our jobs, foreign cheaters have ransacked our factories, and foreign scavengers have torn apart our once beautiful American dream.

Here’s a graphic² showing the devastation that was wreaked on the USA over the past 30+ years:

Can you see how relentlessly say, er, Cambodia (49% Trump tariff) and Lesotho (50% Trump tariff) have preyed upon the poor ickle USA, defenceless against their depredations? Me neither. Ditto for Markistan.

Soooo, I think we can summarise the whole Trump tariff thing as follows…

Okay Mr USA. You bought from me and I’ve repeatedly extended you credit because your credit seemed solid, and now the money you owe me is my fault!??

This is just the start

I’m sure that many people will immediately point out that after Trump’s back-down on Wednesday 9 April, the stock market almost clawed its way back up again. Look at the following graph:

So everything is almost OK, right? Even superficially, this misses the point. And the point is this: Why did Donald duck? It’s obvious that he was pressured—but this simply passes the buck. Why?

Because of the bond market, naturally. It’s difficult to express in words how I feel about that rise in the Dow Jones on 9 April. But let me try. Here goes …

Nope. It’s not because he’s merely deferred that crippling 10% tax on every trading partner. And it’s not because of the yuuuge tax he’s slapped on Chinese imports, starting a massive trade war and turning off the flow of inexpensive goods into the US. And killing US soybean farmers again. It’s far bigger than that!

It’s the bond market. The foundation on which pretty much all world trade is based has been based is the long-term US bond market. We’re talking ten and twenty years+, here. That’s the reason why everyone has given the US credit for so long. Trust.

Ordinarily, when the stock market slumps, people find a warm, comforting refuge in bonds. This means that the issuer of the bonds can offer lower returns—a lower long-term bond rate. People will still snap them up, so why offer more?

But not this time. Not only did the US dollar drop unexpectedly, but here’s what happened to the long-term bond rate:

(Image from CNBC)

That should not have happened. That is the weirdest thing you’ll ever see! Ever.

What does this tell us? It tells us that people with long-term perspective and deep pockets just don’t trust the US any more. They’re the ones who bullied Donald’s camp followers into persuading him to buckle. But the damage is done. And the US is going to pay. Bigly.

In a matter of months, Donald Trump has utterly destroyed long-term confidence in the US.³

Irreparably.

A true confession

I started writing a very different post a week ago. I started with “Orange Monday”—Monday 7 April 2025—and explored the strange oscillation there, and what it might mean. I think that post might be a lot of fun, but this one seemed more natural.

I now have two choices:

1. Resume my interrupted discussion of Deming and Japan

2. Put up the Orange Monday post.

Which would you like me to do? I’d value your thoughts in the comments.

My 2c, Dr Jo.

Postscript

We finally settled on Deming…

1. Bo vs boo. I know. Using the word “Bo!” instead of “Boo” is trailing my coat. I’m simply begging you to point out that I should have used the latter. But in my defence, here’s a quote from 1870:

To be able to say Bo! to a goose is to be not quite destitute of courage, to have an inkling of spirit, and was probably in the first instance used of children. A little boy who comes across some geese suddenly will find himself hissed at immediately, and a great demonstration of defiance made by them, but if he can pluck up heart to cry 'bo!' loudly and advance upon them, they will retire defeated. The word 'bo' is clearly selected for the sake of the explosiveness of its first letter and the openness and loudness of its vowel. [Walter W. Skeat, "Cry Bo to a Goose, "Notes and Queries," 4th series vi Sept. 10, 1870]

2. If you extend that graph from Our World in Data back a few more decades, it’s pretty similar. Poor America!

3. Footnote: The loss of confidence in the US is beautifully illustrated by Paul Krugman on Substack. Read his post and weep.