I obviously cover many economical things in the ordinary course of business, but I try to reserve the sufficiently out of place or in the weeds stuff that is not time sensitive for updates like this one.
It's hard to say how the economy is growing on such a short time scale...
But if you're referring to the fact that the stock has taken a huge dive over the past couple weeks, it's due to recent elections suggesting Milei might find himself out of power soon. It's the possibility of losing Milei that is hurting the economy.
Yeah, I'm being a bit facetious. The linked tweet was from June. It seems that Zvi writes these roundup posts over many months, they could probably do with a once-over before publishing to make sure the information is still timely.
These election results directly follow from Milei's economic policy, so it's not orthogonal. One of the requirements of a successful economic policy is that it has to be able to reproduce itself by winning elections.
Perhaps that's because in addition to an impressive 1st quarter growth, the Milei government wasn't corrupt, the unemployment rate wasn't rising, the social the safety net wasn't gutted, and wasn't begging for a bailout from the U.S. (just months after needing another bailout from the IMF) because they know their economy isn't sustainable and is contracting. A tweet that is void of important context and out of date isn't evidence unless one is working backwards from a conclusion.
Likewise when quoting an inaccurate claim about increasing the capital gains tax causing people to leave the UK when the timeline doesn't support the claim, it's working backwards. Oh and given the Laffer curve has proven to be rather inaccurate given all those tax cuts that somehow never produce the predicted increase in revenue demonstrate, using it to try to score points is especially silly. When reality disproves the theory, you shouldn't cling to the theory because you want to believe it.
Is easy to grow 5% if you made the economy crash 10% the year before. Argentina was having an economic rebound after Milei destroyed the economy right after taking office. Saying that is growing at 5.8%, withouth any context, is just wrong. Also, from the beginning of 2025 the economy slowed down and in the last several months is in a big crisis again. Milei needed a IMF rescue and now needs the US Treasury to rescue him again.
So, basically Milei destroyed everything, got the economy into a big hole, then it kind of rebounded, and now is again at the same place where it was. But having to pay back more debt from the IMF and the US. Milei failed spectaculary. Is disgusting.
Capital gains are an especially elastic tax base because, unlike other forms of income, you get to decide when (and in some cases if) the taxable event occurs. This is downstream of two features of the tax code:
1) taxation upon realization (rather than accrual)
2) the step-up in basis at death, which strongly encourages those who are optimizing over generations to never sell appreciated property and instead financing consumption through other means
Without reforming either of these two features, the revenue-maximizing capital gains rate will be much lower than that of income more broadly.
Appreciate the link, since I cba to unfold Twitter threads. It's very weird to hear Patrick speak - distinctly unpolished, yuge nerdy distractible autist energy (my god, the irrelevant tangents...), not at all how I imagined he'd sound given his elegant writing style. Good on Russ for being able to digest the infodump into a comprehensible summary.
I'd forgotten he has his own podcast! But this makes me meaningfully less likely to check it out, unless his performances there are significantly different. Will stick to BAM.
(Zvi sounds...about exactly as I'd expect in audio form, which I guess is a sign of good calibration? Similar thing though, where half the fun is in the reading. Oral vs written word culture, or something.)
After much thought on this topic, I have come to the conclusion that Patrick is correct, but I don't think he does an amazing job of getting his point across.
My key concern is the one which Russ outlines very clearly in this conversation at 52 minutes: "do rewards credit cards, which have higher interchange fees associated with them, push up the prices for everyone, punishing in particular poorer people who can't qualify for them?". Even if this was in no real sense "redistribution", it might still be making the world a worse place for the poorest in society, and result in a situation where the margin on poorer customers was higher than that of richer ones (after netting off rewards).
Ultimately I don't think this is an important or significant effect, while I can't rule out the possibility that there is some truth to it. In supermarkets in particular, the subsidisation of the price sensitive (often selecting loss leaders or near loss leaders) by the price insensitive (selecting higher margin products) is a much more significant effect.
If we assume that rewards cards holders probably spend far more on higher margin products, the cost of the rewards (via interchange) is almost certainly completely subsumed by the higher average margin of those customers.
“If we assume that rewards cards holders probably spend far more on higher margin products, the cost of the rewards (via interchange) is almost certainly completely subsumed by the higher average margin of those customers.”
I listened a while ago, but I think this is exactly one of the points Patrick made in the interview.
And good point on supermarket prices—hadn’t thought of that.
Alex Recouso is wrong about CGT in the UK. HMRC data show capital gains tax receipts fell ~18% to £12.1bn, a drop of ~£2.7bn year-on-year in 2023-2024. For those not familiar with UK tax years these run from 6th April one year to 5th April in the next and so these figures cover a year in which there was no CGT increase. The drop is most likely a normalisation after two unusually high years. There's no reliable data on "high-net-worth individuals and families are leaving the country".
The Uber market doesn’t seem like a mature duopoly to me. For one, Waymo is a real competitor. They’re bigger than Lyft in San Francisco and expanding.
The other factor is that it’s a two-sided marketplace and the other side of the marketplace has a second market tied in, the Uber Eats / DoorDash food delivery market. If Uber lowers its margins in taxi services to expand its base of drivers, that helps it in food delivery. Especially since Uber is profitable it makes sense for them to be trying aggressively to expand on all of these fronts.
> The obvious follow-up question is why is there not epic capital flight by every dollar that isn’t under capital controls?
There's a reason so many wealthy Chinese buy property abroad, why so many people buy gold in Hong Kong, etc.
The reason people don't successfully evade the capital controls at a larger scale, is basically that if you are have enough capital to have the ability and motive to do so you're prominent enough that the party knows where you live, and has enough insight into your company via party cells to notice where the money is going. Then you get a friendly chat about how your kids are doing at school and the importance of investing in China's future.
One of Dwarkesh's China podcasts talks about this -- forget which one, but basically the gist was that people aren't allowed to do the capital flight thing because large transactions are all monitored by the party. And also (the guest predicts) if there is a significant enough political disruption, such that the regional financial monitoring offices are not being constantly supervised by HQ, there *will* be massive capital flight instantly, because you can bribe the local guy to approve your transaction. The only thing holding it back is that the local guy knows he'll get caught.
Regarding the comment about living "paycheck to paycheck", I live very comfortably in a situation like this, with SS and a government pension that covers all of my ongoing operational needs. I have investments that can be liquidated for capital expenditures like the new roof I had installed this past month, and for any "emergencies".
The real problem is that the definition of "paycheck to paycheck" is too amorphous. Like many other terms used in the social "sciences". They can be twisted to accommodate whatever point you want to make. So, these "sciences" are not really science, but just social studies, and just as useful.
I’ve seen some definitions that would mark you as living paycheck if you responded that to cover some unexpected expenses you would use a credit card or liquidate stocks.
I agree with you. I do use a credit card to pay for unexpected expenses, but I pay off my credit cards, in full, every month, so I should not be included in this class, which implies a lack of liquidity. And liquidating investments to pay for expenses is exactly what insurance companies do every day, to pay claims for unexpectedly large events. They may keep piles of cash around for the events they can reasonably expect to occur, statistically, but they also liquidate investments, too. And they also buy "insurance" from re-insurers, who also have to liquidate assets to pay for exciting events.
This is my point. The definitions of words and phrases that are used to perform statistical studies are manipulated to create the support needed to justify the political position. It is why all these social science "studies" need to be taken with a pinch of salt. A very large pinch. They are not "science" because the actual measurements of the variables can be manipulated.
Are capital gains down in the UK due to wealthy people fleeing the higher tax rate, or due to the UK economy getting wrecked by a whole host of bad policies starting from Brexit?
As well as giving preferential loans to state owned or favored enterprises Chinese banks have also been made to give loans to local governments at extremely preferential rates, which are rolled over perpetually.
Also links to the housing debt bubble. Lot of local governments used loans to build, and used land as collateral, so if the valuation of property drops you have a problem. (Again so far they've solved this by ordering people not to lower prices. This is presumably a sustainable solution).
Two MLB games a year was once minimal. My mom had an uncle who would take her to baseball games in the 1950s - he went to games all the time, and he was quite low income. He had worked for years for low pay in a religious order as a working brother, until he hurt his back and couldn't work, so they kicked him out. At the time of the baseball games he was working as an elevator attendant, a job that, needless to say, does not exist today.
When you cite an alleged fact like "Argentina outgrows China", you should link to the actual source of that fact, and not to a retweet of a tweet by New Hampshire's self-proclaimed "ambassador to Argentina".
The Parfit's Diner effect (cf. Hershel of Ostropol) is definitely a large part of why my family has a strong tipping norm, which I continue to uphold...presenting oneself as The Sort Of Person Who Tips Nicely, and ideally also following through with the actual tip at the end, just seems to lead to better trades all around. Tipping poorly/not at all for shitty service is perfectly compatible with this view, since if you gave every expectation and assurance that you'd tip nicely and yet got poorly served anyway...well, GIGO! And it doesn't mater for future interactions, since why would you want to dine out again at a place with crappy service? That's half the point. It does raise the question of what's the proper temporal tipping distance though. Would pre-tipping insure even better service? Delivery services and some "order at the counter/kiosk" things have that option, but it's hard to say what effect this has on performance...since the very optionality means someone who doesn't get a pre-tip might give lousy resulting service, even if one intended to reactively tip bigly later, conditional on good service.
Also nice to get confirmation that there's such a thing as a Shitty Tippers In Restaurants list among industry insiders. Even if it's only at the higher end, that's where the bulk of tips are anyway (probably? I'd assume it roughly follows the same distributional curve as income tax?), so Working As Intended. Markets aren't efficient without clear and consistent feedback!
> In other trade is good news, Argentina under the crazy libertarian Melei is now growing at 5.8%, faster than China.
How's it doing the past week?
https://finance.yahoo.com/quote/ARGT/
It's hard to say how the economy is growing on such a short time scale...
But if you're referring to the fact that the stock has taken a huge dive over the past couple weeks, it's due to recent elections suggesting Milei might find himself out of power soon. It's the possibility of losing Milei that is hurting the economy.
Yeah, I'm being a bit facetious. The linked tweet was from June. It seems that Zvi writes these roundup posts over many months, they could probably do with a once-over before publishing to make sure the information is still timely.
These election results directly follow from Milei's economic policy, so it's not orthogonal. One of the requirements of a successful economic policy is that it has to be able to reproduce itself by winning elections.
uggh you'd think their country would care more about money. It's called "Argentina," not "Paupertina"
Perhaps that's because in addition to an impressive 1st quarter growth, the Milei government wasn't corrupt, the unemployment rate wasn't rising, the social the safety net wasn't gutted, and wasn't begging for a bailout from the U.S. (just months after needing another bailout from the IMF) because they know their economy isn't sustainable and is contracting. A tweet that is void of important context and out of date isn't evidence unless one is working backwards from a conclusion.
Likewise when quoting an inaccurate claim about increasing the capital gains tax causing people to leave the UK when the timeline doesn't support the claim, it's working backwards. Oh and given the Laffer curve has proven to be rather inaccurate given all those tax cuts that somehow never produce the predicted increase in revenue demonstrate, using it to try to score points is especially silly. When reality disproves the theory, you shouldn't cling to the theory because you want to believe it.
Thanks to Dave92f1 for the link. The 5-year chart shows that drop to be minor noise.
Is easy to grow 5% if you made the economy crash 10% the year before. Argentina was having an economic rebound after Milei destroyed the economy right after taking office. Saying that is growing at 5.8%, withouth any context, is just wrong. Also, from the beginning of 2025 the economy slowed down and in the last several months is in a big crisis again. Milei needed a IMF rescue and now needs the US Treasury to rescue him again.
So, basically Milei destroyed everything, got the economy into a big hole, then it kind of rebounded, and now is again at the same place where it was. But having to pay back more debt from the IMF and the US. Milei failed spectaculary. Is disgusting.
Capital gains are an especially elastic tax base because, unlike other forms of income, you get to decide when (and in some cases if) the taxable event occurs. This is downstream of two features of the tax code:
1) taxation upon realization (rather than accrual)
2) the step-up in basis at death, which strongly encourages those who are optimizing over generations to never sell appreciated property and instead financing consumption through other means
Without reforming either of these two features, the revenue-maximizing capital gains rate will be much lower than that of income more broadly.
Re: credit card redistribution—Russ Roberts interviewed Patrick McKenzie shortly after that piece in The Atlantic came out. Very edifying (most of which is in that X thread). https://www.econtalk.org/inside-the-mysterious-world-of-credit-cards-with-patrick-mckenzie/
Appreciate the link, since I cba to unfold Twitter threads. It's very weird to hear Patrick speak - distinctly unpolished, yuge nerdy distractible autist energy (my god, the irrelevant tangents...), not at all how I imagined he'd sound given his elegant writing style. Good on Russ for being able to digest the infodump into a comprehensible summary.
I get the impression you don't enjoy his speaking so much, but if, like me, you do, you should check out Complex Systems, Patrick's podcast.
This is actually how I discovered Zvi.
I'd forgotten he has his own podcast! But this makes me meaningfully less likely to check it out, unless his performances there are significantly different. Will stick to BAM.
(Zvi sounds...about exactly as I'd expect in audio form, which I guess is a sign of good calibration? Similar thing though, where half the fun is in the reading. Oral vs written word culture, or something.)
After much thought on this topic, I have come to the conclusion that Patrick is correct, but I don't think he does an amazing job of getting his point across.
My key concern is the one which Russ outlines very clearly in this conversation at 52 minutes: "do rewards credit cards, which have higher interchange fees associated with them, push up the prices for everyone, punishing in particular poorer people who can't qualify for them?". Even if this was in no real sense "redistribution", it might still be making the world a worse place for the poorest in society, and result in a situation where the margin on poorer customers was higher than that of richer ones (after netting off rewards).
Ultimately I don't think this is an important or significant effect, while I can't rule out the possibility that there is some truth to it. In supermarkets in particular, the subsidisation of the price sensitive (often selecting loss leaders or near loss leaders) by the price insensitive (selecting higher margin products) is a much more significant effect.
If we assume that rewards cards holders probably spend far more on higher margin products, the cost of the rewards (via interchange) is almost certainly completely subsumed by the higher average margin of those customers.
“If we assume that rewards cards holders probably spend far more on higher margin products, the cost of the rewards (via interchange) is almost certainly completely subsumed by the higher average margin of those customers.”
I listened a while ago, but I think this is exactly one of the points Patrick made in the interview.
And good point on supermarket prices—hadn’t thought of that.
> Up until a few months ago, liberal and conservative Americans held pretty much the same views on free trade.
My new accelerationist stance is that liberal support for free trade reaches 100% during trumps turn and Milei conquers the US
Podcast episode for this post:
https://open.substack.com/pub/dwatvpodcast/p/economics-roundup-6
Alex Recouso is wrong about CGT in the UK. HMRC data show capital gains tax receipts fell ~18% to £12.1bn, a drop of ~£2.7bn year-on-year in 2023-2024. For those not familiar with UK tax years these run from 6th April one year to 5th April in the next and so these figures cover a year in which there was no CGT increase. The drop is most likely a normalisation after two unusually high years. There's no reliable data on "high-net-worth individuals and families are leaving the country".
Zvi just loves an anti-UK story.
The Uber market doesn’t seem like a mature duopoly to me. For one, Waymo is a real competitor. They’re bigger than Lyft in San Francisco and expanding.
The other factor is that it’s a two-sided marketplace and the other side of the marketplace has a second market tied in, the Uber Eats / DoorDash food delivery market. If Uber lowers its margins in taxi services to expand its base of drivers, that helps it in food delivery. Especially since Uber is profitable it makes sense for them to be trying aggressively to expand on all of these fronts.
> The obvious follow-up question is why is there not epic capital flight by every dollar that isn’t under capital controls?
There's a reason so many wealthy Chinese buy property abroad, why so many people buy gold in Hong Kong, etc.
The reason people don't successfully evade the capital controls at a larger scale, is basically that if you are have enough capital to have the ability and motive to do so you're prominent enough that the party knows where you live, and has enough insight into your company via party cells to notice where the money is going. Then you get a friendly chat about how your kids are doing at school and the importance of investing in China's future.
One of Dwarkesh's China podcasts talks about this -- forget which one, but basically the gist was that people aren't allowed to do the capital flight thing because large transactions are all monitored by the party. And also (the guest predicts) if there is a significant enough political disruption, such that the regional financial monitoring offices are not being constantly supervised by HQ, there *will* be massive capital flight instantly, because you can bribe the local guy to approve your transaction. The only thing holding it back is that the local guy knows he'll get caught.
Regarding the comment about living "paycheck to paycheck", I live very comfortably in a situation like this, with SS and a government pension that covers all of my ongoing operational needs. I have investments that can be liquidated for capital expenditures like the new roof I had installed this past month, and for any "emergencies".
The real problem is that the definition of "paycheck to paycheck" is too amorphous. Like many other terms used in the social "sciences". They can be twisted to accommodate whatever point you want to make. So, these "sciences" are not really science, but just social studies, and just as useful.
I’ve seen some definitions that would mark you as living paycheck if you responded that to cover some unexpected expenses you would use a credit card or liquidate stocks.
I use credit cards for nearly every expense!
I agree with you. I do use a credit card to pay for unexpected expenses, but I pay off my credit cards, in full, every month, so I should not be included in this class, which implies a lack of liquidity. And liquidating investments to pay for expenses is exactly what insurance companies do every day, to pay claims for unexpectedly large events. They may keep piles of cash around for the events they can reasonably expect to occur, statistically, but they also liquidate investments, too. And they also buy "insurance" from re-insurers, who also have to liquidate assets to pay for exciting events.
This is my point. The definitions of words and phrases that are used to perform statistical studies are manipulated to create the support needed to justify the political position. It is why all these social science "studies" need to be taken with a pinch of salt. A very large pinch. They are not "science" because the actual measurements of the variables can be manipulated.
Are capital gains down in the UK due to wealthy people fleeing the higher tax rate, or due to the UK economy getting wrecked by a whole host of bad policies starting from Brexit?
Yes the headline is fake news. The decline is from before the increase!
As well as giving preferential loans to state owned or favored enterprises Chinese banks have also been made to give loans to local governments at extremely preferential rates, which are rolled over perpetually.
Which contributes to the stagnation as all the banks give savers meagre interest, and it's very difficult for everyone else to get credit. So far it hasn't come to a head, since they can just order the banks not to call in the loans. (Useful summary; https://www.atlanticcouncil.org/blogs/econographics/sinographs/beijing-extends-and-pretends-to-deal-with-its-mountain-of-local-government-debt/ )
Also links to the housing debt bubble. Lot of local governments used loans to build, and used land as collateral, so if the valuation of property drops you have a problem. (Again so far they've solved this by ordering people not to lower prices. This is presumably a sustainable solution).
Re revenue, I read that stripe counts subscription fees as revenue even if the company offers 100% discount? Surely that can't be right?
Two MLB games a year was once minimal. My mom had an uncle who would take her to baseball games in the 1950s - he went to games all the time, and he was quite low income. He had worked for years for low pay in a religious order as a working brother, until he hurt his back and couldn't work, so they kicked him out. At the time of the baseball games he was working as an elevator attendant, a job that, needless to say, does not exist today.
Without checking, I would guess that this study is not looking at the cheap seats.
https://imgur.com/gallery/ue07fAW
When you cite an alleged fact like "Argentina outgrows China", you should link to the actual source of that fact, and not to a retweet of a tweet by New Hampshire's self-proclaimed "ambassador to Argentina".
Here is an actual source that contextualizes it.
https://www.reuters.com/world/americas/argentinas-economy-expands-58-q1-year-on-year-2025-06-23/
The Parfit's Diner effect (cf. Hershel of Ostropol) is definitely a large part of why my family has a strong tipping norm, which I continue to uphold...presenting oneself as The Sort Of Person Who Tips Nicely, and ideally also following through with the actual tip at the end, just seems to lead to better trades all around. Tipping poorly/not at all for shitty service is perfectly compatible with this view, since if you gave every expectation and assurance that you'd tip nicely and yet got poorly served anyway...well, GIGO! And it doesn't mater for future interactions, since why would you want to dine out again at a place with crappy service? That's half the point. It does raise the question of what's the proper temporal tipping distance though. Would pre-tipping insure even better service? Delivery services and some "order at the counter/kiosk" things have that option, but it's hard to say what effect this has on performance...since the very optionality means someone who doesn't get a pre-tip might give lousy resulting service, even if one intended to reactively tip bigly later, conditional on good service.
Also nice to get confirmation that there's such a thing as a Shitty Tippers In Restaurants list among industry insiders. Even if it's only at the higher end, that's where the bulk of tips are anyway (probably? I'd assume it roughly follows the same distributional curve as income tax?), so Working As Intended. Markets aren't efficient without clear and consistent feedback!