dcolkitt

> but it seems like the largest player in the market is trying to crush it in order to save the dollar.

The US dollar index is now stronger than it's been at any time pre-pandemic since 2002. If anything the Fed is most likely trying to do the opposite, weaken the very strong dollar, as it did in the early 2000s and 1980s.

Savings are not anywhere close to "dwindling". There are still huge excess savings accumulated during the pandemic. Relative to 2019 baseline ($1.4T annual), Americans accumulated about $2.2T in excess savings.

The 2022 shortfall relative to baseline is about $800B. Even if this continued it would take another 2 years to burn off excess pandemic savings. And that doesn't even count the sizable gains in home equity and investment portfolios over the period.

> Younger people not using FB though seems pretty real.

We've already gone through one cycle of "young people are abandoning Facebook" with millenials and Snapchat. And it turned out that the thesis of Snapchat as FB killer was widely overblown. Snapchat failed to really expand into other demos the way early Facebook did, not to mention a lot of those Millenials aged out of it and migrated to Instagram.

Now we're hearing the same story but with Gen Z and TikTok. But it's hard to see a compelling reason why TikTok will wind up a different story. Seems very unlikely that my kids pre-school will eventually wind up uploading to TikTok the way they do Facebook.

My guess is the youth vanguard will always use whatever the new hotness is in alternative social media platform, because keeping grownups out is the point. But Facebook (and Instagram) seem like they're firmly entrenched as the base layer of civil society.

At current price, Facebook doesn't have to remain a growth stock. It's trading at a 15 P/E multiple. For comparison Proctor and Gamble is trading at a 25 P/E, Coca Cola at 26 P/E, and Comcast at 33 P/E. For Facebook to go up (at least relative to prevailing American equity valuations), it just has to remain a blue chip marquee name and not go into terminal decline.

Unless you have an insanely optimistic viewpoint on socialism, it's very unlikely that every laborer will gain if capitalism ends. Even Lenin recognized the need to cap the wages of the highest paid skilled laborers.[1]

[1]https://www.marxist.com/wage-differentials-under-lenin-and-t...

There are many class groups that have a potential grievance against capitalism. But pretty sure that FAANG engineers are not one of them.

Even if your business is entirely unaffected a recession, it may still be sensible to do a light layoff before things turn. Especially if you've just gone through several years of an extremely tight labor market.

The reality is most of the big tech companies have had to staff under pretty unfavorable conditions over the past few years. That's meant either scraping the bottom of the barrel in terms of the less talented. Or ballooning TC packages. If you think that all of a sudden it's going to become a buyer's market for employers, then you probably want to free up some spots today. Especially among your weakest performers or your most overpaid.

It more has to do with tech companies having expanded headcount way more than other industries in the past few years.

The reality is that tech cos overbites since market conditions were extremely ebullient in the tech sector circa 2021. A large fraction of those new hires never had anything productive to even do.

Also pretty disconcerting that they're doing this at the same time that they're pumping billions into OpenAI (a company with near zero revenue) at a $30 billion valuation.

MSFT stock is barely moved on the announcement. Doesn't necessarily seem like the market is rewarding them.

There's 1.1 million physicians in the US, and about 12.8 million worldwide. Almost all of whom are making substantially lower salaries than their American counterparts. Many more than 50% lower. Even if 10% were willing to immigrate for a huge wage boost we could replace every doctor in America multiple times over.

The issue isn't a shortage of residency slots, it's that we don't allow doctors with foreign residencies to practice in the United States.

The physicians in Spain and South Korea and Costa Rica are really good. Arguably better than American doctors based on health outcomes. Many would jump at the chance to earn high American wages. There's no reason a South Korean doctor with ten years of experience in the world leading hospitals of Seoul shouldn't be allowed to practice medicine in the US just because he didn't do a residency in rural Alabama

Most Tether users either aren't in North America and/or are directly holding on-chain. Add up all the US and Canadian centralized exchanges, and accounts for much less than 10% of Tether's outstanding float. Now I certainly don't know if Tether has enough reserves to redeem 100% of their supply, but I'm almost certain they can redeem 10%.

Two issues here with the comparison. First Coinbase isn't just an exchange in the sense that the CME is an exchange. Coinbase is an exchange, a clearinghouse, a custodian, a broker dealer, a prime brokerage, and a user platform. When you want to trade on CME, you don't just go to CME dot com, open an account with them, and have them hold your money. There's layers of other service providers, with CME itself basically being a B2B platform that also offloads much of the banking and custody to other institutions.

Second, Coinbase has ambitions beyond just being an exchange and that's driven a lot of hires. It runs Coinbase Ventures, Coinbase Cloud, an NFT platform, develops tooling (e.g. Coinbase Wallet), and contributes to crypto protocols. A lot of the recent hiring was driven by these "other bets".

Core CPI ex-shelter was negative the previous two months. The way BLS measures shelter is notoriously laggy. High-frequency metrics of shelter inflation (e.g. Zillow data)[1] is falling off a cliff.

The upshot is that while CPI is probably over-measuring inflation today, and we're basically back to baseline, earlier in the year we were probably running closer to 11-12% inflation with the real housing numbers.

[1]https://www.zillow.com/research/rent-inflation-31602/

Don’t think anybody doubts that the current achievements are intellectually impressive. The issue for VCs is how valuable are they from a business perspective. Creating photos of people in places they’ve never been is really cool, but realistically how much money will people actually pay for that?

For AI to be as valuable as current rounds are implying it has to actually be capable of replacing your lawyer, your accountant, your software engineer. ChatGPT is nowhere near accurate enough to do that. It’s the same issue we ran into with super cruise control versus fully autonomous driving. Or speech recognition. Early progress was fast. Last mile in AI is always brutal.

Maybe there’s some argument to say ChatGPT won’t replace your lawyer or software engineer but will make them 20% more efficient. Possible. Though I’m still skeptical. ChatGPT’s unparalleled ability to create really convincing sounding errors is a big hidden negative productivity drag.

So far, when asked what's the use case for ChatGPT/LLMs, the sole answer with clear PMF is marketing copy. Agree it is very good for that. Just as crypto had very clear PMF for cross-border payments and stablecoins in middle income countries without dollarized banking. The problem is neither of these use cases is sufficient to justify peak valuations. (OpenAI is currently raising at $30bn.)

Like crypto, there are other potential use cases on the horizon for LLMs. The problem is the tech is still far too clunky, unreliable and impractical for the foreseeable future. LLMs are subject to a devastating hallucination problem, where they confidently report clearly erroneous data. And there is no solution in sight. Marketing copy is one of the sole major use cases, where random inaccuracies is mostly not catastrophic. But you're never going to trust LLMs to write code or legal contracts or technical documentation or even your own emails until the hallucination issue is drastically improved.

That leaves us with a few other niches that are, frankly not very monetizable. ChatGPT is very good at writing student essays, but students don't have very much money to spend. Makes a fun chatbot, but people are unlikely to pay much of anything to have a chat companion.

Like crypto, AI is a case of really intellectually interesting technology that nerdsniped a lot of smart people. Like crypto, AI has a few small (in terms of market size) use cases where there's clear PMF. Like crypto, there are potentially a lot of other much larger use cases that the nerdsniped smart people are imagining. Like crypto, these large use cases are going to require a lot of fundamental improvements to the tech itself before they become feasible. Like crypto in 2021, AI valuations in 2023 are treating those large scale use cases as imminent rather than long-term speculative.

(Note this isn't a statement about which tech has a brighter long-term future. It's still more than possible to believe (as many on HN do) that AI will be transformative over the next couple decades and crypto is vapor. But the point is AI valuations are implying that it's going to be transformative over years not decades, and it's pretty clear that's not the case. There are parallels with crypto in 2021, where regardless of the long-term promise, the short-term returns simply cannot support the valuations.)

There have been a few valiant attempts to build crypto exchanges with modern technology and matching engines. Generally the problem these run into is that, while the algo traders and market makers love them, they have no competitive advantage in attracting retail order flow. Trading firms have no interest in just trading against each other, so without retail flow nothing happens.

Yeah, as an exchange FTX was pretty terrible. The matching engine was notorious for being high latency, unreliable, and having all sorts of undocumented issues. Every algo trader there will tell you horror stories about trying to cancel an open order, getting a cancel confirmation, then getting notified that the order was executed. Sometimes seconds later.

What really gave FTX a competitive advantage was its cross-margining system. It was super easy to post collateral in any supported asset, and all open positions were automatically margined against each other. So if you were doing something like pairs trades between spot and futures, you didn't need to constantly monitor and manually move capital between the silos. In hindsight though it's not clear if this was possible because of better engineering, or just because Alameda was internalizing all the risk.

So there are a couple of reasons that things wouldn't necessarily converge to equilibrium, and they're fairly well documented in all the asset classes you mention.

Foreign stocks: There's a well-documented effect called "home-country bias"[1]. Basically, investors tend to buy much fewer international stocks and are overly concentrated in their local domestic stocks. There's all kinds of reason for this, but they mostly seem to center around regulatory barriers (typically harder to open a fund to invest outside the country), political (major pension system are encouraged to invest at home for patriotic reasons), and reputational (losing money overseas tends to make the asset manager seem more reckless).

Real estate: Most governments heavily subsidize real estate from a combination of tax advantages and cheap credit. Look at how easy it is for the average Joe to buy a house with 400% leverage, no margin call, no capital gains on sale, and he gets all kinds of tax credits. Which means if you are getting those advantages than it's rational to invest in real estate, but if you look at raw returns real estate tends to underperform because investors with those advantages are willing to accept lower returns.

Bonds: The disparity between equity and bond returns is perhaps the most studied in all of finance, and is known as the "equity risk premium"[2]. There are numerous explanations, but two major ones standout. First, bond returns tend to be anti-correlated with the general economy. During recessions, when people are most likely to need liquidity, bonds tend to go up whereas stocks tend to go down. Second, many large classes of investors are basically forced to invest in bonds instead of stocks. For example insurance companies can only hold a tiny percent of their reserves in stocks and are required to invest in fixed income products. Similar story with banks, and to a lesser extent pension funds.

[1]https://www.gsam.com/content/dam/gsam/pdfs/common/en/public/...

[2]https://www.yardeni.com/pub/stockmktequityrisk.pdf

Metabolic set points can and will drift over time. So if somebody starts out with a setpoint at 22 BMI, they might naturally be anywhere from 21-24 BMI at any given time. As food in the environment becomes more hyper-palatable and calorically dense, people are more likely to "settle" at the upper end of this range. Over time spending more time at the upper end of one's range will "nudge" up the setpoint gradually and continuously.

But the point is this is a slow and continuous process that occurs over years or even decades. Discontinuously making a big shift in weight, up or down, is basically impossible. Which is why despite very high rates of obesity, it's basically unheard of for someone to gain 50 lbs in a short period of time. Almost all weight gain takes the form of 5 lbs one year, 5 lbs another year, and so on.

So you agree, the overwhelming impact of diet on health is from reducing obesity?

And surely you also agree with the science that we now have drugs that are far more effective at reducing obesity than any diet known to man.

Air pollution seems unlikely as it's been falling for decades, and America has much lower levels than Europe. Microplastics are a perennial boogey man, and probably a genuine ecological issue, but there has no been mountains of studies done and documented health effects are virtually non-existent. Livestock and slaughterhouse conditions are certainly not good from an animal welfare perspective, but we've had similar practices for 100+ years and obesity wasn't a major issue in 1960. Pesticides in food products have been substantially reduced since the EPA changed guidelines in the 1990s, and yet obesity has exploded since then.

GLP-1 agonists were discovered in the 1980s, and there are already statistically significant population samples who have been using them to treat T2D for decades. Any serious common side effect would have been already discovered from the large-scale clinical trials that have been repeatedly conducted.

Non-obese people who eat McDonalds and drink Coke are unequivocally healthier than obese people who adhere to the healthiest diets. There's absolutely zero academic research that disputes that obesity (and exercise) by far is the largest risk factor for diabetes, heart disease, etc.

Try to find any major research showing either diet has a significant health impact on either 1) non-overweight individuals who remain non-overweight; or 2) has a health impact after controlling for changes in weight/obesity/body fat. It's virtually non-existent. Perhaps there are small differences on the margin, but all dwarfed by the impact of obesity.

Again this shouldn't shock us, because we know hunter-gatherers across the globe eat tremendously varied diets that come in all sorts of extremes. Yet, modern diseases of affluence (heart disease, diabetes, etc.) are virtually non-existent in these populations. The only thing they have in common is a lack of obesity and non-sedentary lifestyle.

You haven't controlled for body weight changes though. Nobody doubts that diets (if adhered to) can reduce weight. So if biomarkers improve relative to an alternative diet, the question is it because A) the Med diet is more effective at weight loss, or B) the Med diet has benefits outside weight loss. The research you linked is interesting, but confounded by weight loss. Research that has directly tried to control for this has generally found that the health benefits of the Med diet are mostly or wholly attributable to the mediating impact of weight/fat loss.[1]

This is a critical question, because GLP-1 agonists like semaglutide are far more effective at weight loss (with much lower dropout rates) than any diet known to mankind. So if the benefit of the Med diet is mediated by weight loss, who cares in the age of GLP-1 agonists? It's like arguing about the fastest hot air balloon after the invention of the jet engine.

[1]https://www.semanticscholar.org/paper/Obesity-Mediates-the-A...

It really is not a frame of mine. We know this because GWAR genome studies find that the SNPs most correlated with obesity are expressed in the hypothalamus, which has nothing to do with higher thought.

Trans fats are indeed unhealthy, but 1) it's been nearly a decade since they were basically removed from fast food, and 2) they don't contribute to obesity.

As for HFCS, there's basically zero evidence that they're a major contributor to obesity versus other types of sugar. (In fact, honey consumed by the Hazda has almost the same balance as sugar as HFCS). We have multiple points of evidence to counteract this. 1) Mexico uses substantially less HFCS compared to other types of sugar, but Mexican obesity rates are as high as the US. 2) HFCS, and sugar in general, consumption has declined over the past 20 years. Yet obesity continues to increase. 3) HFCS was widely consumed in the 1980s, when the obesity rate was de minims.

Modern nutrition science has found that each person, at any given time, has a "metabolic setpoint". This is basically a target weight that their endocrine system is trying to keep them at. There's a feedback loop, where below this weight hunger hormones will nudge them to eat more and metabolic hormones will lower caloric expenditure (lowering body temperature, sleeping longer, fidgeting more). And vice versa above this weight.

Most people will have a +/- 25 lbs window, where the feedback loop is minimal. But once they step outside this window the nudges become increasingly strong. This is why it's generally easy to lose 15 lbs, but really hard to lose 50 lbs. Also why many can lose weight under very targeted lifestyle interventions, but rapidly snap back as soon as they relax even a little bit. Finally it's why the vast majority of normal weight people don't need to count calories, their metabolic setpoint intuitively matches their appetite to their caloric needs.

"Eat fewer calories" is technically a solution, in the same way that "stop drinking alcohol" is a solution for an alcoholic. The problem is someone who has an obese-level metabolic setpoint will literally be ravenous 24/7 once they get down to healthy weight. If you haven't experienced this yourself, try fasting for 48 hours, putting out a bowl of chips on the table in front of you, and resist the urge to eat even one while you try to do something else. That is what it's like to be an obese person trying to lose a large amount of weight.

GLP-1 agonists work by short circuiting the hormones that regulate hunger. It's basically the equivalent of moving your metabolic setpoint 100 pounds lower. The need to explicitly count calories no longer exists, because your appetite will naturally limit itself well below the point of caloric expenditure.