The need for GPUs during an unprecedented spike in demand for chip fab capacity couldn't have helped.
I think the basic model will eventually work out, though. The bandwidth is there. The compute has to be cheap enough that the biz model works by just taking the retailer's cut of the title sale + maybe a tad more. The tad more can come from one-time hardware sales and maybe better negotiated cuts of the sale from the studios. I would never ever have purchased a copy of any AAA title without Stadia and similar services. I think the same is true for almost all Stadia purchases.
But Google's failure here makes sense. It's a low margin game. google sucks at low margin games.
I had a month off between jobs and used a week of my evenings off to play through red dead.
I don't particularly care about losing the license, because I just wanted to play the game through. Haven't touched it since then.
Stadia's ideal market wasn't "real gamers" -- those folks will buy rigs. Their ideal market was people like you and me, who don't have the time or interest to justify purchasing a gaming machine but still want to play through a AAA title or two every once in a while.
> This sounds like post-hoc, anti-intellectual rationalization.
How is this anti-intellectual?
If you want, I can formalize as a game the problem of choosing business/product strategy in a competitive market with a continuous flow of imperfect information. I can then use ideas from controls to establish some upper bounds on what can be inferred from a continuous flow of information. I can then use that result to prove an impossibility result about the game. I can even tweak assumptions to get bounds on probability distributions which infer we'd be better off flipping a coin or whatever.
I'm not going to do the work, because intuition is almost always enough to identify these situations, but it's absolutely clear to me that results like this obviously exist and correspond to many real-world situations.
> Good judgement requires data.
It used to be that insisting on data-driven decision making was a hard pull. Now it's the opposite. Insisting on data where data cannot possibly provide enough signal to make a decision is the new form of anti-intellectualism. IMO.
> How do you explain a non-negligible part of middle of America who voted for Obama voting for Trump?
The labor movement always had a socially reactionary underbelly. My hypothesis: it wasn't bad economic times that activated Obama-Trump voters. It was, rather, good economic times that triggered those voters to flip from labor-first to identity-first. The tension was always there. This is why the Tea Party fizzled and why Romney couldn't activate the switch -- reactionary blue collar folks knew intuitively that those movements weren't "on their side". But by 2016 a bombastic billionaire (also obviously not on their side) could be excused because the economy was good enough to at last put identity first.
I grew up in a suburb that flipped and my wife is from a mid-sized non-metro city that also flipped. Obama Trump voters are substantially all of our social circle.
A lot happened, but honestly, "youtube's algorithm and super effective conservative media" is probably the best explanation in over 1/2 of the roughly 3 dozen anecdotal cases. It's just pure identity.
Some blame Obama for outsourcing and the GFC, but that's mostly noise. First of all, the major factory that closed wasn't outsourced. It closed because the company was wildly mismanaged, and was not replaced with a foreign factory. Also, this happened five years before the GFC and Obama's election. Most of the folks in our circles remained employed throughout that GFC and all of them did very well from 2010 on-wards.
They're currently all doing very well financially; far better than the late 20s/early 30s city dwellers whose conversations I overhear in the coffee shop. Red America can afford single family homes with big yards, saves for retirement, and just generally lives a very comfortable life unimaginable in blue cities. With jobs in law enforcement, trucking, and in hospitals which don't require even an associate's degree. It's really the American dream -- graduate from high school, six weeks of training, and you're making enough to afford a house and kids and retirement within a couple years.
The resentment and utter hatred of liberals is visceral and real. Pretending it's economic is bullshit cover. Tell someone sharing a bedroom in NYC that their counterpart with less education, who works fewer hours, and lives in a SFH with a pool hates them because of the unfair economic spoils of the former.
Also, the economy was doing extremely well in 2016. Economics is now top-of-mind, but it didn't even come close to motivating them to vote for Trump in 2016. That was pure identity politics.
> economic conditions dictate that within a few decade, there will be no restaurants in major cities.
No. Restaurants will continue to exist. Sit down in particular. In cases where you are paying for an experience; the economics will get worse, but they will continue to exist.
Fast food restaurants will even continue to exist. But the latter only with substantial automation. This has already happened at the front of the house -- kiosks and apps are the "happy path" ordering interfaces at every McD's in a major city. I'm merely projecting that the most cost-sensitive segment of the food services industry will push that automation into the kitchen, the checkout line, and a lot of the administrative work that happens at branches.
Someone working low-paid hourly wage work can probably make $20K. A few siblings + their parents = 5 people = $100K. Stagger availability schedules so there's always someone home with the kids, and share vehicles.
Suburb of top-tier cities? No. Cheapest suburb of a midwestern city? Absolutely.
Again, not saying it's reasonable. Just that it's possible.
You can fit 3 generations in a single family home in an American suburb of a mid-sized city, and half the world's population would either (a) consider that a QoL improvement or at least (b) put up with it for a while to remit back home.
For all the doom and gloom in the USA, it's still an incredibly rich country.
To be clear: I'm not making a moral statement here. This is a statement of fact, not a statement of ethical preference.
Fast food is pricing itself out of the market. I think it's a great example of a sector that failed to innovate and is now incapable of providing a product worth buying at prices that are profitable.
McDonalds et al. probably should've been investing a lot more in automating their kitchens. The fact that a huge percent of locations -- and all drive-thrus -- still use humans for the ordering process is, similarly, inexcusable.
Unless the US massively increases its low-skilled immigration quotes, robotics firms will be the FAANG equivalents of the 2020s-2030s.
> I'd rather not base my picture of the entire economy around a single retailer
Or retailers in general. Retail spending is down because services spending is up -- people are going to the beach and buying plane tickets instead of buying TVs and patio furniture.
Regardless of whether we're in a recession, retailer numbers aren't a good indicator for the duration/depth/type. We spent all of 2020 hearing that demand was pulled forward. Well, it was pulled forward from somewhere, and now we are there.
> What's going on with other retailers?
Best Buy missed by a lot; sales down by over 10%; forecast was a 1% contraction.
Amazon tonight. Always complicated because of AWS, digital content, and now a substantial advertising business. If you want the retail details you have to go past the headline numbers.
Target in mid-August but expectations are similar.
My guess is that none of those companies will be hit has bad as Walmart because (1) Walmart was just particularly badly mismanaged, and (2) these other companies are just different in kind (particularly Amazon -- they could eg suffer retail losses while beating estimates on Advertising/Cloud)
>>>GDP has long been criticized for being a poor metric for the economy. But it's always been correlated enough with overall economic sentiment that it was Good Enough.
>> yes by MMT supporters that want to ignore the classic model of economics in favor of monetary manipulation for political purposes
Huh?
Discussing the utility of and then issues with GDP has been a staple of Macro courses for decades. E.g., [1], but I'm really serious: every single macro textbook for as long as macro textbooks have existed has had some version of the question "Discuss the strengths and weaknesses (i.e., problems) of using GDP data".
> This isn't true, major retailers have been reporting falling demand.
As OP said,
>> and services.
> Wal-Mart
WalMart's issue is that they have a massive inventory/demand mismatch. They already have the headwind of a shift from goods to services, and then on top of that they also massively mismanaged a shift in consumer preference within goods. I have a feeling that they are also feeling the squeeze from Amazon. Many households treat Prime as a fixed cost but the 20 minute drive out to Walmart is a real expense that can be substituted with Amazon purchases. Tonight's earnings will be interesting.
Compare to eg Visa [1].
I expect that the "last hoorah" spending of this summer will grind to a halt in the winter and by Q2 2023 we'll be able to see a massive decline in consumer spending during Q4 in particular.
But we aren't there yet, at least in aggregate, because consumers are spending like mad on services.
Descriptions of economic conditions are obviously politicized. This was probably always true to an extent, but it really took off during the early Obama years.
That said... my wealth manager has been and still is waffling when I ask him if we are in a recession. It's NOT just political; I don't even know my wealth manager's politics and I have absolutely no doubt he puts fiduciary responsibilities first and makes fact-based assessments.
Unemployment is low and the trend is mostly flat. Housing is doing fine. Consumer balance sheets are strong. Are we entering a recession? I think so. But there are a lot of "but"s which really do impact how smart and impartial people are thinking about where and how to deploy capital.
When I ask my wealth manager "are we in a recession?", he basically says "We might be in a recession. We might not be in a recession. Either way, we should not invest as if this is a typical recession."
On the other hand, we just went through a once per lifetime pandemic and a once per lifetime consumer stimulus package. Supply chains are wrecked and we're in a period during which international trading patterns could massively change. Meanwhile,
* China is still purusing zero COVID.
* There's a land war in Europe with Russia effectively cut off from the Western world.
* German manufacturing, once a workhorse, is being crushed by energy costs
* much of the developing world is swimming hard against the current just to stay alive.
I don't think we are in a situation analogous to 1947, of course, and I do think we are in or at least heading into a more-than-technical recession. But "irrelevant because that only happens after once in a lifetime extremely disruptive events" is a pretty... odd... take in 2022.
Hah! An Accounting prof told me that I should double in Accounting because CS would be outsourced. I said something like "well, maybe, but CS is going to automate a ton of Accounting jobs either way, and the accessible Accounting jobs that continue to exist will require some programming. So I will choose my job maybe being outsourced over my job definitely being automated."
I'm now on a few committees at the college and mentioned this conversation to him. He remembered it, even though it was may years ago. He laughed and told me that's basically what happened -- for a few years they had very low in-field placement rates because there are not enough personal tax prep jobs to go around and the larger accounting firms demanded programming skills for entry positions. The Accounting major now requires multiple CS courses and he encourages his students to double-major in CS.
On that note, I'm bearish on SWE compensation and bullish on CS research compensation over the next 10 years. For basically the same reason. We've been massively under-producing (good) CS PhDs because industry pays so well and the standard advice is "only idiots get PhDs when industry starting pay for folks who can get into good phd programs is 150K+".
Basically, "zig when everyone else is zagging" is pretty good early career advice.
> I was concerned about rampant outsourcing for some reason.
Because this concern was pervasive in pop culture / career advice columns during post dot-com. Fear-mongering about outsourcing is probably one of the primary causes for the huge run-up in software development salaries in the 2010s.
I always enjoyed programming (like, from 3rd grade onwards). Basically, from my first line of BASIC, I was hooked.
I realized I was built for computer science during my first industry internship. The job was in probably one of the most interesting corners of the industry, but I still found the day-to-day slog of software development was boring and tedious. The only part of the job I enjoyed was interacting with the researchers, who were working on really interesting and difficult problems that I had no idea how to solve.
I figured I would die of boredom within 5 years if I became a software developer, or have to switch domains every few years and get my stimulation from the domain details. But I really enjoyed the computing part, not the applications. So I became a computer scientist instead of a software engineer.
This Sounds like financial engineering. It is possible the people involved were incompetent, either individually or as a group. But this sounds like an incentive problem. Competence, misdirected.
> Branded as a success
I fear you aren't cyncical enough.
Was it a filure? Did the company's balance sheet improve? Did the company's stock price increase? Did the spinout attract a load of investment for its (pre-destined) success and turn that into a rich revenue stream?
Markets -- for both debt and equity -- are mostly idiotic machines on short timeframes. Your ability to generate free cash flow puts a floor on your price. Others' imagination and available liquidity are the only ceilings.
Additionally, these types of spin outs can behave like generational wealth. They may be incompetent idiots, but who cares? Daddy's money is more than you can raise and mommy's rolodex is richer than you could possibly imagine.
I think the basic model will eventually work out, though. The bandwidth is there. The compute has to be cheap enough that the biz model works by just taking the retailer's cut of the title sale + maybe a tad more. The tad more can come from one-time hardware sales and maybe better negotiated cuts of the sale from the studios. I would never ever have purchased a copy of any AAA title without Stadia and similar services. I think the same is true for almost all Stadia purchases.
But Google's failure here makes sense. It's a low margin game. google sucks at low margin games.