How tech’s defiance of economic gravity came to an abrupt end
economist.com
economist.com
There are sectors of tech that will absolutely struggle. Gig-economy companies won't ever meet their valuation. Negative PE ratio startups are realizing that hyper-scaling is not as infinite as they believed, and that means their earnings might never catch up to their price. Social media companies are finally facing stiff non-American competition (Tiktok). Lastly, it is nice to see the crypto scams be snuffed out. But, people have been calling all 4 of these out since before Covid.
Now, the biggest movers in tech seem to be in a healthy position. AMZN*, MSFT, AAPL and GOOG sit comfortably at the same average PE ratio as the S&P500. Those are value stock numbers for companies that are all leading innovation while having solid unit economics. They are front runners in areas that are their biggest risks (AI, Silicon) and their current offerings are essential tools to everyone globally. AFAIK, the adoption rates for cloud compute and mobile-silicon devices is only going to go up from here.
[AMZN] - AMZN does the weird profit-reinvesting thing, so they report lower margins on the highest revenue across tech. Thus the lower PE ratio. See it as you wish.
Advertising revenue. That’s what Facebook and Google were supported by.
And it has been a zero sum game in a race to the bottom for years. They just didn’t diversify much, same as Russia or Venezuela with their oil economies.
Most companies built around zero-sum games are eventually going to stop growing as you run out of other people’s money. But that’s OK — they are already HUGE!
The real problem is the public stock market (also a zero-sum game, mostly) creating a bubble and pushing them to extract rents from their ecosystem harder and harder. It’s fine to raise money in order to grow, but investors end up reselling shares for more and more money and represent a class of people who extract rents from a multi sided market.
(By contrast, true utility tokens pegged to the dollar that can be sold to actual customers and are a way to raise money that is more economically sustainable and doesn’t create speculative bubbles.)
It's sort of like the difference between coupons in the local paper, versus coupons that your ad man hands out next to the front door of your business. Everyone who gets a coupon was coming here anyway.
The dark side of this is how your business can suddenly disappear from view if you stop buying into the Web-Ad-Hive-Mind. That's way worse than the coupon guy hanging outside the front door - that's more like someone in fake construction vests putting up cones around your parking lot.
[1] Giving them something that looks like magic powers to individual businesses, particularly those that aren't particularly computer savvy. "Holy smokes!! 95% of the people who Google reaches went to my pizza place! They must be soooooo good at convincing people my pizza's the best!"
Soooo… ads provide substantial value to businesses?
Yelp likes to do that, for instance. Alternatives to ads include:
Referrals / word of mouth
Organic rankings
In a way, ads are similar to speculators pumping a token they think will eventually justify the hype by recouping the speculative spend. There are diminishing returns the bigger the ad spend gets.
The only difference is that there is no secondary market to keep reselling ads and scalping them higher — or is there? But there is certainly the element of a large supply of greater fools to try out ad spend on various speculative ventures.
What was last major innovation at Google.. chrome? 20% of company drives profits while 80% is a decorated R&D lab chock full of benefits, upset about cutbacks in their free oatmilk lattes, even though they’re not coming to office anyway.
these guys will double their attrition targets if not have full blown layoffs. ads drives their profits, but search has been in decline for years now.
google right now is a bloated bureaucratic whale that needs to hit the gym and get in shape. Step 1 is to get off the damn couch.
and no, google doesn’t have anything significant in this space. “Innovation” behind search was to slowly increase no. if paid ads in your results, driven by need to increase revenue and keep the rest of the 80% of biz afloat.
LAMBDA and PaLM are literally the answer to Chat-GPT. Google is their closest competitor.
I do agree that they will lose their moat, and soon Google search will just be another tool among many.
For that matter, wasn't Google collecting masses of GIS data and doing intense self driving car research? Where is that product?
Serious q - do google have so much money that they are doing tons of research and then just forgetting to make a product out of it?
The NYTimes and the rest of the media misread between the lines on this story.
Google aren't worried about competition. They are worried about losing the war with SEO.
ChatGPT showed them a glimpse of the SEO endgame. It upsets the balance between search ranking/boosting and SEO. Whatever adaptations they put into their boosting algo, GPT-style AI can eventually learn. Once learned, it can be subverted.
Fundamentally search boosting is at a disadvantage, because SEO has (blackbox) access to the boosting algo by feeding content to the crawlers and running searches. However the booster doesn't have access to the SEO algo; in fact it doesn't even know (beyond a guess) how many discrete SEO algos it is fighting, or which sites are optimized by which algos. So there is great asymmetry there.
That is why the fire alarm was pulled.
This has been a major part of the FAANG strategy: crush competition by giving people who might work for or become potential competitors an offer they can't refuse. 80% of the company ostensibly produces minimal value, but the real reason you're giving them huge compensation and oat milk lattes is to make sure they don't work for, or found, a Google competitor at some point in the future.
This has led to a perpetual feedback loop: FAANG company gives their engineers big stock grants, which ensures no competition in their markets, which increases the value of their stock due to no competition, which allows them to hire more engineers with big stock grants, etc...
We're suddenly entering an era where FAANG stocks aren't continually moving upward. This is largely unprecedented territory; these companies have never experienced an extended downturn since this became the industry hiring strategy.
If the recession continues through 2023, I would not be surprised to see these companies abandon this strategy and only retain necessary talent. Would not be surprised to see 50%, even 75% staff cuts at Goog and Meta specifically.
The second largest country in South America is Argentina. It has a land area of about 2.8 million square kilometers (1.1 million square miles). Argentina is located in the southern part of South America and is bordered by Chile to the west, Bolivia and Paraguay to the north, Brazil to the northeast, Uruguay and the South Atlantic Ocean to the east, and the Drake Passage to the south. The country has a diverse landscape, ranging from the Andes mountain range in the west to the Pampas grasslands in the center, and the subtropical rainforests in the northeast. The capital of Argentina is Buenos Aires, which is known for its rich cultural heritage, vibrant nightlife, and tango music and dance.
1. Brazil: 3,287,957 sq miles
2. Argentina: 1,073,518 sq miles
3. Peru: 496,225 sq miles
"On two occasions I have been asked, 'Pray, Mr. Babbage, if you put into the machine wrong figures, will the right answers come out?' I am not able rightly to apprehend the kind of confusion of ideas that could provoke such a question." "
ChatGPT is a variant of the GPT (Generative Pre-training Transformer) language model that has been specifically designed to generate human-like text in a chatbot-like fashion. One of the main advantages of ChatGPT is its ability to generate natural and coherent responses to user inputs in real-time. This makes it well-suited for use in chatbots and other interactive applications where users expect a quick and personalized response.
In contrast, search engines like Google rely on pre-existing content and do not generate responses in real-time. While search engines can be very useful for finding information that already exists on the internet, they are not designed to generate original responses to user queries.
Overall, ChatGPT's main advantage is its ability to generate human-like text in real-time, which makes it well-suited for use in interactive applications such as chatbots."
There are many Google searches that are low effort queries for advice. ChatGPT can provide realistic answers to those in a way Google never could. I've received compelling advice on how to have a telephone conversation with an awkward relative and how to handle a complaint in customer service. Each response seems genuine and well thought out. In contrast Google might spit out a few lines extracted from a possibly relevant website but then I'd have to trawl through a million ads and back story to know if it actually was relevant. Even if the chat bot returned an irrelevant response I could clarify my question and it would remember the context.
Transformers were invented at Google, which is the underlying tech on which GPT is built. I would say that counts as innovation.
See: https://en.m.wikipedia.org/wiki/Transformer_(machine_learnin...
Google has done interesting stuff. Ideas, research papers, R&D in AI, Glass, Stadia, leading in all kinds of niches… Google Docs with real time collab editing for example. I’m not arguing against that, and to respond on that thread is a straw man.
But they can experiment all day and come up with all kinds of cool stuff but at end of day you need to monetize and make money. Not grow endless headcount, not build more and more empires of a professional managerial class who care little about actual value and more about their fiefdoms and promotions.
To make money, Google depends entirely on X and unless they quickly course correct, they risk massive disruption to their biz.
Google lost their way. The culture stagnated into one of corporate bureaucracy, red tape, and employees feeling persecuted because the contractor responsible for chocolate crepes on Thursday didn’t show up one week.
They have to start getting in shape. There’s no way around it. Yeah they’ve entrenched themselves but unless they change it will be a slow painful decline. Focus on a few things and do them extremely well instead of 50 different things that ultimately go nowhere. And figure out how to get off the ads lifeline
There’s alphago and alpha star etc. What about Waymo (splitting the company into pieces doesn’t make it a different company)? Since chrome they have made a lot of hardware too. Not to mention the dead attempts at innovation which didn’t pan out like stadia
AMZN: IIRC, AWS is the majority of their profit. This is web infrastructure; all the negative PE ratio startups and gig economy companies going under will hurt AWS. They have similar risks to GOOG (see below).
MSFT: Does anyone have a clue how MSFT is making money? The last few Windows updates have been free, and I would guess they are buttressed by Office 365. How they convince IT departments to adopt this buggy, slow, expensive software, is beyond my comprehension. What’s the sales pitch? “Hey, have you ever wished you could pay every month for Office? It will have similar functionality to Office from 20 years ago, and you can pay us every single month!”
On second thought, their Azure services probably have similar profitability and risks to AWS; maybe that’s what is keeping them afloat.
AAPL: At least Apple produced physical products, but we have seen their products, e.g the iPhone 14, hurt by supply chain issues and lower demand in the face of a recision and high inflation. They’re not innovating; the iPhone 14 was mocked for being the same as the previous model. If frugal consumers go by the wayside, who is driving profits? The (now-laid-off) tech employees who make enough to afford their luxury products? Maybe some, but less than at present.
Apple has made efforts to grow their SaaS revenue the last few years and that is prone to risks from lower consumer demand. A budget-minded consumer will not be paying for all the Apple services when they are hurt by layoffs.
GOOG: Ah, my favorite. Let’s set aside the legion of small companies they bought to fuel growth and decrease competition, e.g. Nest, Fitbit, Waze, YouTube, etc. Corporate Alphabet is producing the majority of profit by ad revenue. What happens when demand for advertising decreases due to lower consumer demand? They’re also trying to enter the cloud space, but are far behind AWS and Azure.
There are also broad risks facing tech company. The main ones which come to mind are data privacy and other regulations, anti-trust actions.
If future events do not favor big tech, those companies could have a long way to fall.
I think (and perhaps I'm very wrong!) we'll settle back into a 2018, 2017 tech industry minus companies like OpenDoor rather than a early 2021 euphoria
This has yet to happen. Unemployment is still quite low. The music is just starting.
As long as that narrative remains strong, nobody will dump everything. New investment might stall, but no panic in the ranks.
The real panic will set when the pivot does not happen. Or if it does happen and inflation comes roaring back, leading to a longer and deeper rate hike.
It’s rent seeking. The internet does not need landlords.
Technology is great, discovery is great; politically correct economic traditions of enabling a minority that can email 20-30 minutes a day and fuck off to the beach with their friends was a huge problem.
The rest of society will not tolerate austerity for them and privilege for prince and princess billionaire heirs as generational churn reshapes political power.
Our statistical tools are curious indeed but the wrong tools for management of living, breathing, beings, and humans have a long history of belief in false prophets that we barely just consciously began moving away from (Vint, Gates, Jobs; they did not invent computing or even transmitting coded signals as electrons through a medium). It’s a hard sell our leaders steeped in a less tolerant, post war shell shocked, Cold War paranoid era, huffing leaded gas fumes, are of “the right stuff” to still dictate our agency.
I could turn out to be wrong but id bet there’s many more waves of cascading pain before before things turn other way. Turning other way doesnt mean we immediately go back to excess hiring, empire building, lack of focus, free uber rides and oat milk lattes.
will be long while before market turns in favor of employees.
- I think we will experience a massive deflationary bust. The inflation narrative is over, it’s all about a potential recession now.
- By Summer 2023 we will experience YoY deflation and the Fed will cut then, or even sooner in anticipation to that.
- We may see hikes pausing as soon as February, and cuts starting in the summer as a result of rapid disinflation (which may turn into deflation). Deflation is a lot worse than inflation.
- We find out that the bulk of inflation was indeed caused by supply chain bottlenecks. As China fully re-opens, those will be resolved.
- Russia’s war with Ukraine will end (Putin is now agreeing to potentially think about negotiations). Oil prices will ease
You can have stagflation: a recession with inflation. This is the economic boogeyman that literally every central banker has worried about for his or her entire career. Inflation is certainly not “over” - still running extremely hot relative to the last decade. Better than before the rate hikes, but by no means good or over.
(which begs the question: what does it mean to be artificially low? Compared to what?)
This is as red herring as it can be. No remotely serious analysis of this war can take anything Kremlin says at face value.
Ukraine already tried appeasing Russia in 2014 after the last invasion and now they have the current situation so I doubt any decision with actual skin in the game (aka Ukraine) are going to be trying that again anytime soon.
"Reopening", and caused by it pandemic speedrun is already bringing back more supply chain disruptions
> - Russia’s war with Ukraine will end (Putin is now agreeing to potentially think about negotiations). Oil prices will ease
Putin is trying to "win" the war by insisting on talks where they keep the territory. Anyone looking closely knows that's not going to be accepted even as a premise to start talks.
We’re also being told that this willingness to negotiate is a sign of weakness and eminent failure for the Russians so Ukraine shouldn’t negotiate and should instead press their advantage.
In effect we should not negotiate with people who are not-willing to negotiate nor with people who are… which only makes sense if we never intend to negotiate - a position that’s much easier to hold when we’re not paying the price of the failure to negotiate.
Ukraine is defending so… I dunno, I guess they aren’t very motivated to make concessions. I guess the starting point for negotiations for them would be Russia leaves, and then the details of where on the spectrum from “huge” to “massive” the reparations are can be worked out.
If people don’t want to negotiate that’s their prerogative. I’m more interested in the difference between what I’m told by media and reality, and more precisely if there has been a softening of Russias stance I want to see where.
We can influence the Ukraine/Russia negotiation of course, by increasing or decreasing our support for Ukraine. But, since they seem to be willing to continue fighting, decreasing aid would be a pretty huge betrayal.
My understanding of the situation is that realistically the US has a lot of latitude to influence Ukraine, but should not go behind their back and negotiate with Russia. Any diplomacy that routes around Ukraine means their priorities will be given less emphasis. Since they are the ones actually paying in blood and human suffering, we owe them deference when it comes to when and how to negotiate. We’re spending money but it is the sort of spending we can handle for a while.
First, hypothetically is the US were to fight Russia, we’d use a different mix of weapons. So it is putting unexpected pressure on our supply of Javelins, because they are doing the work that would normally be done by, whatever, JDAMs or something.
Also there must be some of what we programmers would think of as oversubscription or deduplication of weapons — I mean if we have like 2 war’s worth of weapons to handle all our competitors and we expend one war’s worth of weapons to handle one competitor, then we suddenly only have one war worth of weapons to handle all the rest. We haven’t expended one war’s worth of resources on Russia but it is worth keeping in mind that this isn’t like, literally zero cost or something like that.
Edit: also another thing that rubs me the wrong way with this description is, it seems to play into the narrative that Russia and the US never could have gotten along. The real hope is that the overwhelming US lead has a deterrent effect on countries that might want to start a war. The best weapons are ones that don’t get used.
Right now the EU is bearing the brunt of the economic costs and they're not wealthy so they'll run out of money sooner. But ignoring the money spent costs; what is being discovered in this war is the efficacy of mass low cost long range drones. It's unlikely to ever be cost effective to defend against this and conflicts settled this way will be immensely destructive. I remember when it was unthinkable to militarize drones to remotely kill people from a long way away, then we did it, and soon it'll be done to us.
Inaccurate. Russia is getting backed into a corner, where they believe the wests goal has been to destabilize and break Russia all along. The more desserts they get more likely they are to use a thermonuclear weapon in Ukraine.
The increasing probability of nuclear war should be sufficient motivation for america to broker a peace deal. Instead we’re sending more weapons to Ukraine.
The media narrative has removed the 20 decades of context on how we got here. Focusing only on what’s happening now, the narrative is that Russia is motivated by some strategy to conquer Europe and any different perspective means the person is stupid or a Putin shill.
I'm still confused when people assert there's any chance Putin will use a nuke in Ukraine.
I'm far from an expert on Russian politics, so I imagine I'm missing something in terms of reputation or intimidation or face saving, but I always figured that Russia wanted to own/control Ukraine, since it's a fertile, resource-rich part of the former USSR (it's a huge grain exporter; consider all the famine warnings since the war began?)
Why would they choose to destroy this thing they want to control? Using a thermonuclear weapon specifically seems really counter-productive. If I stretch I could imagine a future where they deploy some tiny tactical nuke for intimidation/escalation, but same logic applies, I don't think they want to destroy the thing they're trying to conquer, so they'd want to limit collateral damage.
This isn’t the motivation. Putin has been saying since 2007 that NATO expansion is an existential threat to Russia. Now whether we agree with that or not is besides point. They view it that way.
They warned in 2007 Ukraine entering NATO would lead to war, but we basically told Russia that there a has been and they don’t matter. This led to what happened in 2014. Then in 2021, we basically come out in full support of pulling Ukraine into NATO.
What’s happening now is extremely dangerous. Russia can’t win in Ukraine because of west helping Ukraine and their economy is ruined after sanctions. They see this as an end of Russia.
You’re backing them in the corner. This is when rational state actors act irrational. The chance of thermonuclear war is increasing because Russia cannot take on the west in a conventional war.
We had a significant role in how we got to this war, but we won’t ever take responsibility. We pulled out of Afghanistan and now the military industrial complex is making billions through Ukraine.
I'm saying "may" because while Russia clearly has the ability to make Nukes, and ICBMs, that would have required constant heavy, maintained, spending on their military that they clearly didn't do for, oh, tanks, air defense, aircraft, new aircraft designs ... At best their Nuclear arsenal is one generation ahead of what the Soviets had, whilst the West's defenses have advanced 5 generations in the same time. I get that's a gamble, but ... Nuclear weapons require constant re-enriching of the warheads fissile material, and maintaining 7000 warheads requires running tens of gigawatts of nuclear power plants without getting any electricity or heating out of them. Not to mention: you need tens of thousands of highly qualified engineers to do it. Are they doing that, not just lying about it? I mean, I get that they're hiding this, but one reason we don't hear much of it could be that it simply doesn't exist anymore. I've never once read an explanation of how this infrastructure survived the fall of the Soviet Union.
Russia is vastly overplaying their hand. We can be sure of that, given the facts.
Second, the Russian economy will die faster than the European or Ukrainian economy over time. Time is not on their side. Russia is slowly losing the ability to attack even without any war, without anyone destroying them. Europe, and European states will gain more and more ability to attack over time. I'm saying whatever their reason was to attack in 2014 and 2022, they will have more reasons to attack as time goes on. This war won't end just because we force it to, it will simply move.
And all of that is assuming we DON'T move away from fossil fuels. If by some miracle we move either transportation or heating of of fossil fuels, then Russia is done for 10 years before such a move is complete. That will definitely kill their ability to support their military. I mean what do you expect Russia to do? Attack France when ITER declares success? Because ITER, and 100 other similar efforts, also are existential threats to Russia.
Russia has "lost", the empire doesn't exist anymore, and the conflict in Ukraine, even if they 100% win, doesn't change that. If their perception is such that if they lose they will use nuclear weapons, then it's unavoidable, and the conclusion is just the opposite of what you suggest.
The correct course of action if Russia will truly will chose to go out in a nuclear "blaze of glory" or if Putin has the ability to "commit suicide by World War 3", then the only option is for NATO to attack Russia, ASAP. For the simple reason that Russia, the empire, is done for.
If Ukraine is "pressing its advantage", it's to take back the territory that Russia has illegally attempted to annex.
Please stop spreading Russian propaganda.
At the onset of the current war, Russia and Ukraine met in Turkey immediately to begin diplomatic talks. You do realize we intervened and told Ukraine to walk away?
This whole narrative that Putin is irrational, that Russia is just plain evil, and any contrary perspective is “Russian propaganda” reminds me of 2003 when questioning the Iraq war effort would get you labeled as a terrorists sympathizer.
This war is far more nuanced than the US government-backed narrative being pushed in the mainstream media. That itself is propaganda.
And what’s truly embarrassing is people like yourself actively try to shut down any discussion or contrarian views. If it doesn’t conform to your view of the world, it’s propaganda.
"Give me your land and I won't kill you all" isn't much of a negotiation, is it.
> "Give me your land and I won't kill you all" isn't much of a negotiation, is it.
And Russia isn’t trying to conquer Ukraine.
Russia doesn’t want neighboring countries, particularly Ukraine to join NATO. They’ve been repeatedly vocal and warning about this for over a decade. NATO kept that door open and said last year Ukraine entering NATO was a certainty hence the war.
Certainly, it’s easier to ignore the past two decades of context, how we got to where we are now, and just draw Nazi Germany comparisons. No one in their right mind would defend Nazis, and so you can use that tactic to ignore facts and quickly shutdown the debate.
IMO you can decide not to put in the research and go with the distorted NYTimes propaganda, but the Nazi comparisons are just intellectual dishonesty and bad faith.
I've read multiple analyses, very cogent and well-sourced, that concluded that what Putin wants is not just to conquer Ukraine, but to reassemble the USSR, or the Tsarist Russian Empire before it, with him at its head. He sees the fall of the Soviet Union as a stain on Russia and on him personally, and he wants to restore his/its "dignity".
As part of this, he believes that Ukraine is Russian, and that any Ukrainian who does not see themselves as Russian is wrong, and likely either an American-sympathizing traitor, or mentally ill.
"He does't want Ukraine to join NATO" is true, of course, but in the same sense as "he doesn't want to lose his power." He wants to keep both of those things and gain much more. Making it all about NATO is 100% part of the propaganda he has been spreading.
Your so-called context is a strange kind of provincialism that makes everything a result of US actions (and thus sets a blame on them), and ignores history and recent developments in Russia and Eastern Europe in general. Neutrality was tried in 1930s, didn' work, countries got picked off one by one. It wasn't an option anymore in 1990s, EE countries weren't going to make the same mistake twice.
Eastern Europe lost any illusions towards "new Russia" when the first Chechen war started 1994. THAT was the turning point in European diplomacy that your pre-packaged "America is bad" narrative ignores because it doesn't fit the timeline. From how the war was waged, it was obvious that nothing had changed and that Russia was still the same shithole with an inferiority complex that uses genocidal wars against neighbors as a coping mechanism. In 1995, the first Chechen president Dudayev predicted that Russia's ambitions would not be limited to them alone[1], and it was clear to everyone else too. From then on, the clock was ticking: either build strong relationships with free countries (and become one) to acquire as good security guarantees as you can get, or get left out in the cold alone to be mauled by Russia.
As to comparisons with Nazi Germany, I see no reason to soften phrasing anymore, any differences have become superficial. Russia is openly a Nazi country. It has an intense personality cult around its leader, a pseudohistoric expansionist racial ideology of "Russian world" in theory and in practice as it wages a war of extermination against its neighbor, with the goal of destroying Ukrainian statehood and people, ultimately forcing the whole Europe under its influence. Lebensraum, Herrenvolk, Volksgemeinschaft, and so forth - it's all represented. What differences are there left, that they don't build freeways as much? It's a country where prime-time TV, under direct government editorial control, discusses bombing London, drowning Ukrainian children and setting their families on fire - none of it would look out of place in Völkischer Beobachter in 1942.
That shows like this[2] are deemed fit to air in Russia (and even represents the mainstream) tells everything you need to know.
1) Russia: for Ukraine to accept territorial concessions (Donbas + Crimea)
2) Ukraine: for Russia to retreat from all Ukrainian territory, (maybe) except Crimea (meaning they're willing to sit down at a negotiation table without the Russians agreeing to retreat from Crimea)
So the two sides are far apart. Russia is judged weak because the prediction is that without negotiations they will lose all territory inside Ukraine. They are dependent on negotiations, and unwilling to accept the reality that they've lost the Donbas (Luhansk and Donetsk), Kherson, and Zaporizhia, and are unlikely to be able to hold Crimea in the long term.
Furthermore, Russia has consistently failed to demonstrate capabilities they threatened Ukraine with. They claimed to be able to take Ukraine "in 2 weeks, maybe less" (https://www.youtube.com/watch?v=h_RReQj7PpY), they claimed they couldn't be stopped in the Donbas. They have been stopped. They have been claiming for 40 years to have the tightest air defense network on the planet. The air defense network couldn't stop drones (which are unmanned planes, designed to fly slowly), not even with 6 WEEKS warning. They have panicked and deployed additional air defense platforms around Moscow, in other words, they're afraid they cannot protect the Kremlin from Ukraine's military, 450km from the nearest Ukrainian border whilst they cannot (effectively) hit Kiev's presidential palace, at 200km from their borders, and less than 70 from their deployed soldiers in Belarus. This is not the way to show strength.
They are known, even long before this conflict, to constantly lie about everything, to ignore agreements made before. So it's not even very clear what an agreement with Russia would even entail. If they say they won't attack, they can't be trusted to uphold the agreement. So what's the point? There must be a good military reason for them not to attack or there's no point.
Or to put it in negotiation parlance: BATNA for Ukraine is better than the conditions Putin demands to even talk. Nevertheless, these aren't small organizations: whether or not they've agreed to talk high-level, they're constantly talking.
"Potentially think about negotiations" is a long long way from the war ending.
Also Putin's terms are absurd. He wants to keep half of Ukraine while evading persecution for the crimes against humanity the Russian army caused against civilian population.
That's not gonna fly in Kiev, who wants all their territories back including Crimea, and the guilty parties held accountable at the Hague for their crimes, plus war reparations being paid, which is not gonna fly in Moscow.
Historically, inflation has never come down so fast or so easily anywhere. So much of inflation is about belief and psychology. If a "good times are here again" belief takes hold, it's entirely possible that inflation will run amok, again.
Asset bubbles don't necessarily turn out in cost-of-living inflation, as we saw in the era of QE after 2008; it was inflationary for asset values, which kept on climbing, but prices goods and services were stable, remaining under the 2% target rate in the decade between 2009 and 2019.
This can be seen from the fact that companies are posting record profits...while also saying "but we have to raise prices 20%, because of 7% inflation!"
This argument cannot stand until it can answer the question: why are corporations all of a sudden greedy now, and not before? If they could raise prices 20% at any time, they would. They don't, because the market won't bear it. It's nonsensical to suggest otherwise.
Pemex, the Mexican state-owned oil company, made $6B in the first quarter of 2022, but lost money every single year for more than a decade before that. Why didn't they jack up prices sooner to make a profit?
Delta Airlines lost a lot of money after Covid, and has yet to neither reach pre-pandemic profitability levels nor recoup the money lost during the pandemic. Are you saying if they were only greedier, they wouldn't be in this situation?
Even the great devil Exxon Mobil itself had ever-shrinking margins leading into 2020 before losing money for four quarters in a row before coming back into the black. Again, is your explanation that they just forgot how to be greedy for a while?
Historically, the sharp bounceback from the brief but deep COVID recession is anomalous in a lot of ways, and the lagging brief inflation spike caused by the overshoot of stimulus in that anomalous recession is part of that. The causes aren't mysterious and just looking at the timeline of the inflation against past inflations in isolation while ignoring other relevant knowledge is...silly.
I highly suggest you provide an alternative number if you really think the US is in a high job loss recession that's somehow being hidden from the American people.
This happened in the late 70s/early 80s. Volcker made this mistake and had to jack rates a second time leading to a second recession deeper and longer than the first.
The current Fed has referred to this in recent speeches. I don’t think they’ll make the same mistake.
It was the same in 2001, BTW.
In the dot-com crash, a lot of people in software especially left the industry and many never really got their careers back on track or recovered financially. On the other hand in 2008/9, there was belt tightening in the software industry but overall the impact didn't end up being all that great. Remains to be seen if this period will be the former or latter. Of course, individuals can still be very affected even if the overall impact isn't huge.
what a discombobulating way to say "clarifying" :)
Agree. Early 2000's was a nonstop litany of entire companies going under because they didn't make any money. IMHO, a fair comparison would be to look at the same companies now. Many funded, growth at all cost companies are laying people off, but we're not quite to everything (outside of crypto) going under like we were back then.
Looking at big tech as the barometer doesn't make much sense to me since the dominant companies are still making massive profits, but they're just slightly less massive and/or growing less quickly. That's not an end of the world scenario like early 2000's or 2008 felt like.
Job growth was less than it was originaly reported. By a lot. From 1M to just 10K.
https://nypost.com/2022/12/22/biden-administration-overstate...
What exactly is the unemployment rate if it's not ~3.7%?
It’s an insight into how they’re thinking for one. For two, the conditions that led to the current economic situation are unlikely to be present in the future. In many cases those situations can’t be reproduced because they would require highly unlikely situations (another pandemic) or impossible situations like the fed going to negative interest rates or near zero tax rates.
Further, population growth is one solid, bedrock economic growth truism. Some of this could be ameliorated by the US taking a “send us your weak, your poor, etc” mindset to immigration. This is how things were pre-2010, ish.
Our economic system will explode violently in a default cascade without population growth or productivity growth and we haven't been able to figure out ways to consistently increase productivity, which is to say we need population growth to sustain our way of life.
It's why deflation is the boogeyman. Deflation is basically a win in a society with zero debt. Deflation is a humanitarian crisis at 130% public debt to gdp and whatever ~300% total debt to gdp we are at.
But population growth is a net loss for most of us. More plastic waste. More energy consumed and emissions created, more space taken up etc. There's already more than enough of us by any metric except the amount of gdp growth required to keep our economy running.
This isn't my opinion, it's how economics define those terms.
Also, labor quantity can go up or stay flat while total population goes down
I guess that's the case for software-centric platforms. It was not my experience.
I worked for hardware-centric corporations, for most of my career, and became used to having software treated as a "nice to have, but not essential" part of the product. In many cases, my work (and myself) were treated with contempt. I got used to being sneered at.
In my experience, this was a disastrous attitude, because, despite lots of folks wishing it weren't so, hardware, these days, is software.
Software pervades everything, from the compiled silicon on peripheral ASICS and FPGAs, to the firmware that drives said chips.
In my experience, firmware was treated as hardware, and the same rigid, waterfall process was applied to firmware, that was done for the hardware.
Worked great.
Until it didn't.
Software is a drastically different beast from hardware. I won't bother going into the reasons. Anyone with a smattering of knowledge in the area, can list them.
In any case, the hardware folks would treat any attempt to leverage the flexibility that software allows as "cowboy, low-quality, laziness." It was Waterfall, or you were a "bad engineer," and "lazy and undisciplined."
I'm really big on Disciplined software development. That does not make me popular with this crowd. It also does not mean Waterfall.
In my opinion, there's no way to avoid the difficult parts of engineering, but it's also important to be adaptive, responsive, and, dare I say it, "agile."
When we increase the surface area of a definition like you are here it makes words meaningless.
My view of words like technology is they are more like sliding windows, covering what the zeitgeist is classifying. Somewhat like the word 'fashion' or 'fad' aren't limited to any one specific kind of dress or style.
The word 'technology' would be less useful if it always had to be qualified to exclude everything from fire and the wheel up to the transistor?
It seems that with software we’re hell bent on verifying Arthur Clarke’s aphorism wrt sufficiently advanced tech being indistinguishable from magic.
Think language models, for a contemporary example.
Reading and writing are "tech", if you insist on going down this pedantic road; but that's not what most people mean by that term in this context.
Cars have been tech in this finance and investment bubble. Tesla, obviously, but then lots and lots of electric vehicle and autonomous driving startups came in with the whole song and dance of “disrupting the incumbents” and “move fast and break things” and “let’s milk customers forever and ever with subscriptions for self-driving taxis as a service”.
Which is in the process of going poof. (Remember Nikola? No. Good.). Lots and lots of hype about startups and subscriptions, and we’ve ended up with GM having arguably the best autonomous driving tech, and Ford having arguably the most hyped recent EV with the F-150.
They are bought because of the software that runs on it.
Said another way, the dependency graph is bidirectional. Software requires hardware to run. Hardware is of no practical use without software. The fixed quantity is the "use case", NOT the software.
Also, software without hardware to run on, or to write n, is even more pointless than hardware alone. At least the latter can be touched.
That generic software company was valued higher than entire industries that supply components that all hardware depends on. Tech isn’t collapsing. But valuations were and continue to be fuckin nuts for a lot of companies and are coming down to more reasonable numbers, which look like collapses.
But it’s still fueled by hype, so unless a strong enough crash comes, it’ll keep bubbling up.
When coupled with the fact that there are still a huge number of companies which haven’t yet converted to using these tools and purchasing the market leader for a premium makes total sense.
With regards to component supply - companies that are producing unique chips are worth plenty, where as those that are making COTS components aren’t.
- Google was the most popular search engine. Had a dominant position in adTech and YouTube was popular
- Apple became the most valuable company in the US and the iPhone was sucking up most industry profits.
- Amazon was by far the most dominant electric retailer and AWS was taking off (disclaimer: my current employer)
- Microsoft had been the dominant operating system for 15 years and Office the dominant office suite.
- Even Facebook was the dominant social network.
Not one startup has disrupted the industry in the past decade.
AirBnB is probably the only major tech company that has created a profitable large business in ten years.
And how would a search engine company buying out a non profitable video platform that had no means of making money have triggered anti trust action?
In order for these acquisitions to have high valuations, big companies must fear being replaced. It is in VC’s interest to stoke that fear. They do this by the threat of replacement at least as much as through funding for actual replacement.
VCs don’t have to care whether disruption happens, but they do have to care about their IRR, and will say or do anything they feel will with high probability increase their rates of return.
They are bought to be an accretive to an existing business or the acquiring company thinks they have scale advantage to multiply the value of the acquisition.
Another way to put it, that these are “sustaining innovations”.
The highest valuations are not paid for sustaining innovations, but for market access risks, which is what this thread was about. The two can be the same thing functionally, but “sustaining innovations” sounds much better in a shareholder meeting.
Neither LinkedIn or GitHub were going to disrupt Microsoft in anyway.
Post Jobs’ death they bought a black celebrity-driven entertainment company. This was absolutely a brand threat as Apple was now associated with Tim Cook, who is perhaps many amazing things but they do not include cool.
Fast forward a decade and Microsoft recognized the game that was being played, which is that a set of six murky quasi-monopolies attempt to acquire diverse revenue streams and not lose information sources or access to their markets. While LinkedIn or Github may not have been direct threats to any of Microsoft’s existing businesses, if someone else got ahold of them Microsoft would have zero social footprint, which would be a big problem for them, having essentially missed out on search as well.
NeXT was already a failure and was transitioning out of the hardware business. Apple couldn’t make a modern operating system to save its life and was getting crushed by Microsoft.
> Post Jobs’ death they bought a black celebrity-driven entertainment company. This was absolutely a brand threat as Apple was now associated with Tim Cook, who is perhaps many amazing things but they do not include cool.
People aren’t buying iPhones because of a producer that most outside of Hip Hop only knew because he was the producer behind a famous White rapper (Eminem).
> Microsoft’s existing businesses, if someone else got ahold of them Microsoft would have zero social footprint, which would be a big problem for them, having essentially missed out on search as well.
Under Satya, they moved away from Windows everywhere to cloud and Office everywhere.
Azure isn’t popular because of GitHub. It mostly targets stodgy old Enterprise customers that are already on the MS platform. That’s not meant to be an insult. I was a stodgy old enterprise MS dev until 2018 when I started moving toward AWS technologies (where I now work).
So for Telsa's valuation to mate with it's real world ambition, it has be aiming to have it's operations as big as Toyota's, great! Being as big as Toyota would put it's market cap at... oh. Less than it currently is.
Tesla is still valued higher than Toyota, Honda, GM, and Ford combined.
Something is broken.
First and foremost, there's the Exceptions:
Google: P/E is more or less 20, decades already
Microsoft: P/E is more or less 20, for a very long time
There are not actually that high. Compare to BABA (P/E is >200), IBM (P/E >100), JD (P/E >600)
And the reverse exceptions, non-tech with absurd P/E:
Tesla: P/E is 40 (down from ~500 I might add)
Boston Scientific: P/E is >100
and let's just shut up about crypto, because ... there's is a theme. Overwhelmingly the ridiculous valuations are financial companies and "semi-"government companies (meaning protected by government, but not benefitting the people of the country that government governs. Like BABA for example, or before their downfall, Theranos). If Tech becomes the P/E champion instead of "almost-but-not-quite" corruption companies that tend to dominate that, I feel that's a very good thing indeed.
From TFA: “But just as chip production bloomed, demand withered, thanks to falling sales of pcs and smartphones.”
The narrative is apparently backed by data. Where’s the data for your counternarrative
Consumer-side, it's really more about tech maturing. If we take your example of owning a smartphone, it means that most people already have smartphones, and since the yearly upgrades are much more incremental now, people don't need to upgrade as often.
We were running ZIRP and some insane level of QE that’s a bit of a puzzle to even measure. Yeah, equities markets were more than a little overheated.
The notional values may have re-scaled a bit but I’ll go out on a limb and venture that Alphabet is still a great business.
Big tech was even worse than this. Monopoly? That implies a vision and proper focus to achieve business goals.
My take is far less generous. This was one of the biggest grifts in recent times. the professional managerial class obsessed not with their products or services but instead on increasing headcount, building empires, and getting promotions. Office politics and political games played by people who produced little to no economic value.
It’s even with 2 years, it’s up 4x on 5 years ago.
The current machine is a totally broken model though. We have optimized to sell software to each other thats sole purpose is to just make more money on what already exists.
We haven't invented anything to the betterment of mankind in quite a while. I'm sorry, but making software (for example) that makes someone else's mundane job easier is not revolutionary - it's a low effort cash grab.
There is so much potential for tech growth in things the world needs, like energy, climate technology, feeding people, battery efficiency, sustainability. That stuff offers real value.
For example, if you're building another ad delivery platform or PaaS, give your head a shake. It might be a great idea, but we do not need your intellect to solve this problem. We need you to put your entrepreneurial gifts to solving real problems that the world has.
There is room for both - but what I'm trying to say is that if you have the means, please try to solve the hard problems instead. Because we really need someone to step up and 99% of what I see is missing the mark.
Tesla and its CEO is a dumpster fire right now, but EVs are a good example of a hard problem that is valuable.
If technology itself were like a stock portfolio (to a degree it is), what we are building is heavily skewed to junk status. Most of the money is in garbage reinventions of the wheel that are just padding the pockets of people that already have too much.
The web services era needs to die in a fire for now - we need better physical technology / hardware solutions for the next chapter.
> If technology itself were like a stock portfolio (to a degree it is), what we are building is heavily skewed to junk status.
Posit a distribution, while providing no measure.
We had a year of uncertainty and semi stockpiling and then when the recovery steps were clearer a huge amount of pent-up investment happened in the following 12-18 months.
Sentiment is mostly negative right now but deals are still being made, where they make sense.
So the question is what happens to VC-dependent companies - do exuberant times return and they continue their addiction, do they do even deeper cuts, wash out, ...?
By that I mean if something grows faster than inflation plus earnings and it is an inflation hedge or at least the revenue is dependent on inflation and therefore it's valuation grows with inflation, then it must come down at some point and here is the kicker, only businesses that actually manage to sell products and generate revenue are actually benefiting from that inflation. So when faced with higher interest rates, only those business will have an easy time.
So what matters here is nominal revenue and nominal interest, not the actual real interest rate. 4% interest at 8% inflation still has a big impact, even though in real terms it is still negative.
* Hosting a file < 10 GiB via Cloudflare's zero egress fee R2 and sending across the Internet, once for every human costs approximately $3k - https://www.wolframalpha.com/input?i=number+of+humans+alive+....
Distribution costs via Bittorrent are even cheaper! I don't actually how much it costs to run a DHT node, or even who's paying for them today, but spreading a magnet link around is really close to free!
Say we've got this mp4 file that everyone wants. Let's use bittorrent for our digital distribution. So we need to: create a torrent, upload it, and then keep their computer on for a couple more hours? Let's pay $400 for the job. I think it's well within the realm for a clever teenager to do it, or a rando you find off of Upwork or Fiverr, that who knows what they're doing, to finish the first two steps in an hour, so I think that rate is plenty generous. But let's also pay for their Internet connection for the month ($100), and a cheapo laptop to do this work on ($500). This brings our digital distribution costs including the cost of labor, using bittorrent, to be able to make billions of copies to be... $1000?
Even if you pay an engineer to do that job, labor doesn't get residuals, so aren't paid for each copy made, so the cost of labor, whatever it is, is essentially flat. Compared to if you were trying to copy and distribute a physical good, the more copies you make, the higher your costs and $1000 just isn't going to get you far at all.
I'm more interested in what you mean by interest rate though, mind explaining how that fits in here a bit further?
Because what is the point of the whole system? We have one system for two different types of goods. One type is limited in supply because it's physical in nature. The other type isn't really limited in supply because it's digital. Why then should the same kind of money be used to pay for them? How about having two parallel money systems, one of physical goods and one for digital goods. How about for digital goods we eliminate copyright and replace it with a system where we grant every person 864.00 digital dollars (DD) each day, and track what eve check person spends their time watching, then distribute the DDs accordingly. So if I spend 2 hours watching a Disney movie, they get DD 72.00 and they further redistribute those DDs to everyone who played a part in its creation.
Then, the government sets up a system where it will buy DDs for $s, so people will still want to make digital goods, but it becomes a distinct system. This lets us ease into a post-scarcity world - because we absolutely already are there for digital goods - but without having to solve some of the issues that UBI faces.
By default all BitTorrent clients are acting as DHT nodes (if they can receive inbound UDP packets). You can disable it in the options, but most don't (unless for private trackers on per-torrent basis). I don't think anyone actually runs a DHT node like Tor exit nodes (except researchers and search engines like BTDigg).
The cost of reprodcuing a pile of bits is quite unimportant compared to the cost of producing that pile of bits the first time around. This dynamic is poorly captured by supply/demand, and this is where copyright laws have always come in. While they are being abused with absurd expiry limits, they are a necessary part of any kind of money-based society.
Without copyright, for-profit companies would insist on ever more draconian DRM and software obfuscation technologies to replicate similar systems, while liberally taking any piece of code they can get their hands on - whether open-source or from another company with inadequate copy protections.
This basic problem would remain true even if all software and media companies were worker owned syndicalist co-ops.