Hacker News new | past | comments | ask | show | jobs | submit login

I've wondered what the effect our modern digital economy has had on consumer price inflation.

Normally, an increase in money supply would cause consumer goods to increase in price, since more people are able to buy them and there is a limit on how much of any particular physical good is available.

This isn't the case, however, for digital goods. If there are suddenly 100 million new people who want to buy a Netflix subscription, it isn't like we are going to see the price of a Netflix subscription go up because there isn't enough Netflix to go around. The marginal cost for a new subscriber is practically zero, so there should be no price increase caused by a shortage.

It would be easy to see that inflation would be essentially zero if ALL goods people wanted to buy were digital ones... no amount of demand can eat up the supply, since supply is practically infinite.

Of course, in the real world, some goods are digital and some are physical. If you gave everyone $5000, some of that would go to Netflix subscriptions, which wouldn't effect consumer prices, and some would go to buying TVs to play Netflix on, which WOULD cause inflation.

I am curious how much of our current "low inflation even with an increasing money supply" is caused by our increasing spending on non-exclusionary goods.




>The marginal cost for a new subscriber is practically zero, so there should be no price increase caused by a shortage.

Yeah, I don't think that that argument works at all. The price does not increase due to "shortage", it increases due to an increase in consumers' willingness to pay. Going by the Netflix example, if Netflix realizes that not too many people will cancel their subscriptions if they were to increase the price by, say, 1 dollar, they would certainly increase the price.

Consumers' WTP is the reason why digital goods are priced differently in different markets. Many digital goods are sold for much cheaper in India compared to developed countries because the Indian market is much more price sensitive. For instance, Netflix costs only about half as much in India as it does in America.


It's also debatable that there are no costs associated with digital goods. If suddenly Netflix had a surge in subscribers and they doubled them over a short period of time, they'd have to invest in infrastructure to support the extra demand. That would cost them in hardware and human resoursces to handle the extra demand. But yeah, digital services have a better situation at meeting demand than physical goods of which, after the produced amount sells out, you have to wait for more to be manufactured, delivered, etc.


But the surge in revenue from doubling subscribers would (way) more than cover any costs in infrastructure spending. This would not drive any increase in subscription cost, which is purely governed by the competition and content acquisition costs, paired with whatever magic number the major investors/board decides is an acceptable profit margin.


A doubling in subscribers might need a trippling in customer service agents (especially if the new customers are not as good at tech and need more help, which goes along with being a late adopter or if the service quality drops because of the presumed doubled usage, and there's more service requests as a result).

If the doubled subscribers requires doubling the number of Netflix OpenConnect CDN boxes, that would mean current capex, and while the additional revenue might eventually pay for it, there might need to be borrowing costs to get the equipment sooner rather than later. Also, right now is a tricky time to get lots more hardware, so a 2021 node might cost more than a 2020 node, even if they have the same capacity.

All that said, without looking at their investor reports, I suspect they have some margin and cash on hand to make things work and mostly profit. They probably also have a target for spare CDN node capacity, because there's some pretty high variability of peak load on new releases and ISP install lead time can be super long. Also, they do a lot of efficiency work to make sure they can push as much traffic as possible from their nodes.


Yes... those Netflix customer service agents


The infrastructure is physical though, and thus is of limited supply. If you really had to double your capacity quickly, you'd have to take the computing power from someone else, at a cost.

Every digital value chain ends up on something physical.


You have never tried to dig a new or more cables under the sidewalk or into the ocean, do you? What you kids think is "free" is in fact heavily subsidized by other people's money.


Very few fiber optic systems are run at capacity.

Upgrading bandwidth is therefore a matter of new optics and router cards, not new cables.

Furthermore new subsea cables bring down the unit cost of bandwidth.

Far more subsea cables have been decommissioned due to them not being cost effective anymore rather than not working.


That's true but I suspect for all intents and purposes the cost delta is pretty marginal.


> Indian market is much more price sensitive

That’s an odd way to say India is way poorer, and there’s no way the avg Indian can afford to pay a US price for Netflix. People being price sensitive is only an additive effect on top of that.

(India => any developing country)


It might sound an odd way to phrase things, but it was both more fine-grained and more meaningful than your rephrasing.

The average Indian earns less than 4 dollars a day. The average Indian still buys their shampoo in little 5ml sachets that cost 2cents because they can't afford the full bottle even though buying the full bottle would be cheaper in the long run. So, the average Indian definitely can't afford to pay the Indian price for Netflix either.

The average Indian is not the target market of Netflix or of any Internet business. All the talk of India being a market of a billion users is nonsense. Only the top 5 percent, maybe 10, of people in India have the disposable income for them to be a potential target customer for most businesses. For all intents and purposes, India is a market of ~50-100 million customers.

When I say that the Indian market is much more price sensitive, I am talking about this group of people, not the average Indian. The reason Netflix is priced half of what it is in the US is because this group of people are willing to pay that much for Netflix. The average Indian is simply not a relevant concern.


I wonder how that compares to the financial demographics of China? I had the impression they were bringing more citizens up into the lower-middle class, but I might have just uncritically accepted some spin.


China's per capita GDP is over 10,000 USD. Their middle class is already huge.

I think a decent analogy is that Chinese market is approximately the size of USA and India's is approximately the size of Canada.


It is not just about being "poorer", people in developing countries are indeed more price sensitive about technology services. They do not perceive the "value" of the product same as the west does.


The issue with the WTP argument is that Netflix isn't a singular good - there are many replacement goods (Hulu, Disney+, Prime Video) that push down the price via competition. The lack of raw materials makes this competition especially good at keeping prices low.

Now, if the oligopoly within video services (or anywhere else) would conspire to raise prices together - that'd be a different story. We've seen this in other digital goods. The Apple and Google anti-poaching agreements come to mind.


I disagree. I don't think that different streaming services are replacements of each other. It is inaccurate to think of them as mere "video services" (if they were even Youtube would a substitute which it clearly isn't). Their value is in the exclusive rights they have secured for different content. If a streaming service has an exclusive deal with any one of your favorite shows, you are pretty much going to have to buy it.

Speaking for myself, I have bought the subscription of at least four video services in India.


It takes time to produce physical widgets to meet demand, for manufacturers to scale, and then start benefiting from economies of scale.

With Netflix, a million people can sign up immediately, supply adjusts instantly and economy of scale is locked in earlier.

For digital goods you also have huge companies investing for the future. Your local hardware store need to continue to make a profit, whereas Amazon can absorb or push back on cost pressures for much longer.

I agree with OP that this must put upwards pressure on inflation of physical goods even if it’s a medium term thing.


There is much less ingredients in digital products that can be a shortage to give away more digital products. Compared to physical ones where capacities are much more expensive compared to the price of the product, I think the parent comment is basically right.

Of course there is chipageddon now, nothing is without physical. The point is the IP and customer service are the major cost drivers, not the property, the assembly line and the workers.

One more thing is also true: you can scale up/down really fast these days and for a fact at least this won't change your price as even your scale-down risks/impacts are much lower.


>Normally, an increase in money supply would cause consumer goods to increase in price, since more people are able to buy them

Not if the increase in money supply goes from the banks to the already wealthy (as cheap loans), which don't use the extra money to consume, but to invest, buy land, and fund small-competition-crushing rent-seeking endeavours.

Then the money supply increases, consumer goods remain more or less the same price, but some stuff like rent goes up.


We saw what happened in 2008 when the loans went to unsophisticated, underqualified borrowers.


2008 (or rather, the lead-in to 2008) was a different case.

In that case the lenders were the ones who were the problem, and the borrowers were indeed using the money as disposable income (well, at least, to buy a home and other consumer goods on credit).

Whereas today the money goes from the printers to the rich people - not the common folk.


> unsophisticated, underqualified

Like Lehman Brothers


problem were the lenders not borrowers


>I am curious how much of our current "low inflation even with an increasing money supply" is caused by our increasing spending on non-exclusionary goods.

Not much? Based on the CPI weights given by the BLS[1] at least 82.238% of the CPI is from non digital goods. This is based on summing up the top level categories which are definitely not digital, ie. Food and beverages, Housing, Apparel, Transportation, Medical care. If you drill down into the remaining categories (Education and communication, Recreation, Other goods and services) and eliminate non-digital goods from there you can probably get that percentage even higher.

[1] https://www.bls.gov/cpi/tables/relative-importance/2020.htm


Yeah, that doesn't say how much of actual consumer spending goes into the goods that are included in the CPI. Even if the CPI were based 100% on non-digital goods, the percentage of purchases that go towards CPI goods could be falling.


Perhaps, but how much of a difference can that make? It's hard to imagine many people spending $100+ on new SaaS/digital goods because of the pandemic.


That is exactly my question! How much are people spending on digital goods, and what percentage of extra income would go towards digital goods.


> Normally, an increase in money supply would cause consumer goods to increase in price, since more people are able to buy them

This isn't actually true, though it's a widespread belief (quite a number of pundits kept making incorrect predictions after the global financial crisis).

Where the logic goes wrong is that an increase in money supply doesn't automatically translate into higher disposable incomes. (And higher disposable incomes don't automatically translate to higher effective demand, though in practice they usually do if you the increase in income isn't extremely unequally distributed.)


You are right, not all money supply increases are equal. However, I have thought about this more in the context of proposals around UBI or other wealth redistribution ideas. The common complaint about them is around just what you talk about, that distributing wealth to poor people will increase disposable income and cause inflation from not being able to keep up with increased demand.

I wonder how much of that is true in our modern economy. Between our incredible latent productive capacity and the sizable chunk of our economy that consists of non-rivalrous goods, I wonder how much of the increased demand we could absorb without causing inflation to rise substantially. For example, how much would food prices go up if poor people had more money? We can produce a lot of food, and the demand for additional food is not limitless. Some of that money would go towards digital goods like Netflix and video games and other digital goods, which are non-rivalrous.

Of course it would have an effect on inflation, but I am just wondering how much.


FWIW I agree with your thought process here.

Proclamations of certainty around UBI are common, but the only intellectually honest position is that we simply do not have enough empirical data to tell with a useful level of confidence how a UBI would affect inflation.

Also, I believe that the only way to collect that empirical data is to implement a UBI at a sufficiently large scale and see what happens.


There really is no reason why you wouldn’t be able to pair government intervention on the demand side (i.e. UBI) with government intervention on the supply side. Simply introduce quotas about supply (e.g. that shops cannot throw away more than 1% of the food they sell), and/or limits on non-productive assets (e.g. freeze rents & Netflix subscription). I’m sure there are many smarter interventions than I was able to come up in 30 seconds as wel.


Micro-management like that very often fails, though. What do you suggest shops do if they risk crossing the 1% threshold, for example?

I agree that a lot can be done if you're careful about how you do it, but if you aren't careful enough, you may easily make things worse.


I've spent a lot of time in volunteering for the food bank in my area and the local grocery stores generally give food away right before it goes bad. They're already incentivized to do this because we give them tax write-offs.


Welcome to the Soviet Union! They tried exactly that and failed spectacularly at it. Centrally planned economy simply does not work, it's a law of nature.


Government price controls inevitably cause shortages and black markets. It is completely infeasible to enforce a limit on throwing away food.


>if you the increase in income isn't extremely unequally distributed

How is your family's net worth percentage change looking compared to Bezos? How is your salary percentage change looking compared to that?

Sorry for laboring your point. It should be labored. Repetatedly & with emphasis.


We are entering a Post Scarcity Economy. A lot of fiction books write about how this plays out. Regardless of what happens a lot of economic theory becomes less relevant.

https://en.wikipedia.org/wiki/Post-scarcity_economy#:~:text=....


It’s just the opposite. The essential goods needed for survival: shelter is becoming more and more Scarce. 34% of millennials can’t even afford their own shelter anymore and are forced to live with parents. Those who can live by themselves are paying over 40% of their income on shelter. There’s been a pretty dramatic drop in living standards here in the US over the last several decades

The fact that we can now afford ever more borderline useless apps, clothes and electronic gadgets doesn’t help much


Housing prices are not driven by scarcity, they are driven by financialization of our economy.

There are multiple cities in England where population has decreased but house prices increased. During lockdown 700k people left London, but house prices kept going up.


Triple the housing supply.

Prices will go down.


The Midwest US can assure you that they've definitely already tried this. We're literally mowing down every cornfield for housing here (I'm typing this from a neighborhood with thousands of units, 90% of all buildings here were built after 2012). It never works, prices are still at an all-time high, even for units on the market 6+ months or longer.

"Just building more" alone isn't ever going to lower prices enough to make housing affordable for most, no matter how many more units you build.

You have to decommoditize housing, or stop force-preventing natural depreciation, or at least put ownership caps on it, or some other thing in addition to the construction.


You can't say it never works because your definition of "works" is completely off base.

The population is continuously growing. We must build housing as fast as it grows to just maintain parity, and even faster to reduce prices.

Saying building new houses doesn't work to lower prices is like being in a boat taking on water and after throwing a couple buckets of water overboard saying "bailing out water doesn't work" because there's still water in the boat.

Unless you ban immigration and babies, you have to build strictly faster than new people arrive. Wealthily people with multiple houses are not the driving problem behind housing shortages.


There are loads and loads of places where houses are built faster than population is growing, or population is just falling, period.

Prices. Keep. Going. Up.

It has nothing to do with population, it has everything to do with interest rates and mortgate terms.


People can afford a higher price at a lower interest rate, but in aggregate they're getting just as much house for their money. Why is that bad? Perhaps because prices will collapse later when interest rates rise? It doesn't seem like an affordability problem exactly.


Are all those new houses occupied? Where are all the people coming from?


- A bunch are staying un-bought and remain empty. (The bottom of the market is closing in hours, the top of the market is sitting for months)

- A bunch are technically getting occupied, but buyers are having to go way beyond their means to get them. (2008-style, except it's not their fault, you have to live somewhere or die, and you can't manufacture used housing, so it's hard to fault them for taking on risky mortgages. There simply isn't cheaper options)

- A bunch are getting bought and turned into luxury rentals, and those rentals are getting occupied. (This might seem like a win, since the end result is people-in-housing, which should be a good thing. But forcing people into unaffordable rental rates permanently hollows out those people's finances, it's super unhealthy unsustainable housing strategy)

- A handful are getting bought and getting turned into illegal unlicensed hotels. (Permanent "Air-BnB-s").

- A bunch are getting bought and stay empty. (According to local realtors, "second-home" purchases are up 300% since 2019. Generally, this is wealthy or upper-middle-class coastal urbanites realizing the currency conversion between the coasts and the midwest, and using that to their advantage)

Individually none of these are the worst thing or the primary culprit. But added all together, it means that "real people buying houses to sustainably live in", is the least likely scenario for any given home on the market right now here. "Demand" is super high, but most of the "Demand" is kind of fake-demand (demand from finance, demand from investment, demand for vacation -- but not demand for housing to house people).

> Where are all the people coming from?

They aren't coming. House prices are up 250% this decade, but we've only got a ~1.5% YoY population growth rate (for our city) and a 1% population loss YoY (for our state). We aren't a major city, and we've built more new housing units than had actual new population for 5 years straight now.


But isn't this creating a house of cards? Or are we heading towards China, where there are in many cases more apartments than people and yet prices have never fallen in the last three decades.


If a large fraction of the dwellings are truly unoccupied then it is a bubble plain and simple.

If it's just a matter of out-of-state money willing to pay a high price to own houses that they can then rent out for a high price after cornering the market, then it's less clear.


Housing supply hasn’t decreased in london, but demand has.

Housing costs are driven by what people will pay, which is driven by salary, there’s always an option to not pay for housing in london - housing in Stoke is cheap, and not much further commute from Bloomsbury than say surbiton. Places like Luton are cheap and are a very respectable commuting distance, but people will pay far more for a house in Hackney


What people will pay is not driven by salaries, it is driven by how much bank will lend.

If tomorrow the banks offer a 1000-year loan you can pass on to your children, people will take it and single-bedroom apartments will cost 50 million plus.


we shouldn't be treating essential goods as speculative assets


That would require a sufficiently high land value tax or permanent ground leases instead of land ownership.


in 2008 they even built "investment houses" far from any water or electricity supply, with no one living there.


I keep hearing people reiterating this idea, but can you show at least one example where this has worked, anywhere in the world?

Or is it a case of 'it will work this time', just like communism?


California has a lot of cities with 70%+ single family zoning and housing prices are exploding there in terms of monthly payments for mortgages or renting.


I'm not convinced that there's been a dramatic drop in living standards in the US over the last several decades (or that, where there has been, it's not mainly a matter of individual choice).

Think about life in the USA 70 years ago:

Do we think that much fewer than 34% of 20-to-30-year-olds lived in multigenerational households? The average square-footage of new homes more than doubled between 1970 and 2015.

Certainly people spend far more on healthcare, education, and transportation now than then. But aren't those goods and services that were simply not available at all to a large fraction of the populace? 10% rather than 30% of people went to college. Most conditions for which people seek medical treatment now probably went untreated.

So many people in 1950 eked out their existences living in a multigenerational house with a single bathroom and no car. Never traveling out of state or experiencing middle-class city culture. Walking a mile to work or performing farm labor more than 8 hours a day. The world is very different today, but it's not at all clear to me that living a comparable lifestyle is not an option for most people today.


If you believe that post scarcity is possible at all, you might consider that space colonization is also possible, but that brings about new significant requirements and needs that our economy will have to meet. Personally I think our civilization's ambition needs to expand so that happens. If we are still putting the "American Dream" as a goal when that can be built and satisfied cheaply by automation, then our goals need to change. A post scarcity economy would be able to meet today's American Dream easily and with little trouble, and yet we can't shake the mindset that you have to earn it because that makes it mean more or something.


it does not have to be either/or. some people can live a low-impact, low-cost post-scarcity lifestyle, while others can aim/aspire for more abundance


Oh I'm not saying we force people to do space jobs or livings, only that it should wayyyyy more accessible and common than it currently is and certain sectors of the economy will need to support that. It gives us more of what we can't realistically create on Earth, which is more land.

The American dream is a challenge for people now. If we advance to where it's no longer a challenge because the pieces of that are plentiful and cheap and is given to everyone by virtue of existing, then we need a higher challenge because we've essentially "solved survival on Earth" and must now expand. IMO it's way better than just becoming an economy that focuses on producing luxuries or inventing new financial gadgets.


As long as people want things they don't have, there will always be scarcity.

I'll believe the Post Scarcity Economy only when I see it.


From the second sentence of the article

>Post-scarcity does not mean that scarcity has been eliminated for all goods and services, but that all people can easily have their basic survival needs met along with some significant proportion of their desires for goods and services


Do you buy as much milk as you can afford? Most food is way beyond scarcity at this point.


No, but I buy more ribeye steaks and restaurant meals than I need.

I'm of course way beyond fearing starvation, but there is still no real upper limit to my food expenses.


The premise for the linked paper, "The Post-Scarcity World of 2050-2075", that "This convergence of peaking production is likely to lead to an age of scarcity. And yet that age of scarcity is unlikely to herald the end of the world. So what lies beyond scarcity?" makes no sense to me, although I have yet to read the whole paper.

Like, yes, we are depleting the Earth of its resources, leading to scarcity, but then that will end because the Earth will be out of resources, therefore we are post-scarcity and everything is free or nearly so? No sense whatsoever.

I think I can guess what the paper is about, that technology is going to develop that will save us from exhausting resources, but we're already inching pretty damn close to various tipping points that will lead to some really terrible things happening with various ecosystems and things are going to have to get a whole lot worse (and a lot of people are going to die) before we even can attempt to go back to some sort of eventual equilibrium. And tech is nowhere close enough to saving us from a good chunk of it.

So yeah, I guess there could be a 'Post-Scarcity World', but only for the fraction of life that will survive to see it (maybe none of that life being human at the rate things are going).


What I will write probably wont be a popular opinion, but I completly disagree that we enter post scarcity.

Post scarcity perhaps exists in some richest parts of selected countries (California? Hamburg?), but even in those places it is often just an illusion. Roads still have potholes and there are homeless on streets. Schools still struggle with supplies. There are also people who work, but whose work does not allow them to get a real place to live, or health services. Then there are other things that are difficult to measure: for example overworking, or drug addiction, which perhaps dont mean scarcity of goods, but seem to be connected with scarcity of services or scarcity of quality of life.

People from the rich countries consume more natural resources than the few billion in developing, or third world countries. What happens now is already completely unsustainable and causes multiple problems: global warming, loss of natural habitats, dying species...

There are billions of people in Africa or India, who dont consume even 10% of the resources that an average American consumes - but they sure as hell would prefer to get to that level.

I imagine that it is easy to make comments about post scarcity when you live in some mega-rich bubble in California, but this is not true even for USA, not to mention the rest of the world.

People in poor countries dont have cars YET and they sure want to drive those big trucks that drink a lot of fuel per mile. But where will this fuel come from? And where does the CO2 go?

And in my opinion inflation is a result of incompetence, of planned policy by central banks to make the rich richer, while the rest to gets poorer. As much as I dont like bitcoin (multiple reasons, economical, unsustainable in real use..), it is right with one thing: the supply is known in advance and finite. We currently live in a time where money is printed out of thin air to save the too-big-to-fail banks and to pump the stockmarket prices, so rich dont lose anything. All of this at the expense of the middle class. The poor already dont have anything. And the middle class is shrinking: from one side those are the only ones who pay taxes, from other, those are the only ones who dont have real tools to defend against inflation.

I mean, of course you will write that one can gamble on the stockmarket, but weren't saving accounts supposed to be the thing for those who are risk averse? I mean, if someone from middle class gambles and loses you will write that it is their own fault. If they dont gamble - and put money on a savings account (so lose money) - also their own fault.

Currently the central banks create inflation higher than interest on saving accounts, what in my opinion is a very big problem. Since the central banks entered the cycle of bailing out bubbles, more and more money will be created: new and new bubbles will have to be bailed out.. and as I said, at expense of the middle class. Who just get screwed by inflation.


>And in my opinion inflation is a result of incompetence, of planned policy by central banks to make the rich richer, while the rest to gets poorer.

You got this backwards. The central bank is planning to make the poor richer, but as a result of incompetence inflation never hit the 2% target. If the central bank did nothing things could be even worse than they already are.


middle class was an abberation of the post-WW2 era when the world needed to be rebuilt.

We live in neofeudalism.


Quote:

    This article is written like a personal reflection, personal essay, or argumentative essay that states a Wikipedia editor's personal feelings or presents an original argument about a topic. Please help improve it by rewriting it in an encyclopedic style. (January 2021)"


They directly linked to the definition, which is not written like a personal reflection.


Post scarcity only for certain things. Things like living space are becoming scarcer and more expensive.


Living space in rural areas is cheap. Living space near good jobs is what's scarce.


In comes a pandemic induced remote work experiment and global satellite internet thanks to SpaceX.

An unforeseen black swan of a decade.


Not in the U.K. average house prices in rural areas are far higher than in towns, as they are more desirable.


That's true, but they are also bigger and better. A block of flats in the middle of nowhere would be worth way less than one in a city.


>Normally, an increase in money supply would cause consumer goods to increase in price, since more people are able to buy them and there is a limit on how much of any particular physical good is available.

Computers and other electronics are cheaper than ever on a real and absolute basis despite increased demand and increased money supply


Top model iPhones have increased price about 15% per year since launch


Yes, and they’ve also improved by orders of magnitude in terms of processing power, camera quality, etc.

Today’s top-end iPhone probably would have cost upwards of $10,000 in 2015, if it were even possible to manufacture.

As an aside, someone did a back-of-the-envelope calculation and found that the equivalent computing power of a 2014 iPhone would have cost $32 million in 1991: https://www.aei.org/technology-and-innovation/how-much-would...


This is a great example, because if Apple had launched the iPhone SE in 2015 at $10,000 alongside their $199 6S, it would have bounced like a bad joke, and probably only a few wealthy people would have bought them.

So in one sense, yes, tech has gotten cheaper, but in another sense there's a limit to how much deflation can really be stated because you can't say someone buying a new iPhone in 2020 is saving $9000 relative to what they would have spent five years prior. They wouldn't have spent it.

There's also a recent inflationary trend in tech, in that old computers actually slow down due to the growing CPU/memory demands of software and web apps to deliver more or less equivalent value to what they used to.


Yes, but so what? My current iPhone now devotes trillions of potentially-useful cycles to NOPs or processor sleep cycles. Does this mean that $32m of actual value is being created for me?


That’s key, people are spending more on mobile phones now than they were a decade ago. The benefit they get is difficult to quantify on an individual basis, certainly the difference between a phone from 5 years ago have very little difference than one from today, better camera perhaps, but how many people who bought a separate camera in 2016 don’t bother now.


A Netflix subscription buys actors, directors, writers, editors, cameras, microphones, lights, clipboards, trucks, sound stages, hair & makeup, trailers, set decoration, tape, clapper boards, generators, radios, VFX render hours, insurance premiums, producers’ risk, and all the other accoutrements of film and TV production.

There’s a lot of accounting and financial engineering and temporal shifting going on under the hood but ultimately you are paying for real things.


Right, but I was talking about the MARGINAL cost per subscriber... those costs you mentioned are fixed costs, and they are the same whether Netflix has one subscriber or a billion.

The marginal cost per additional subscriber is basically bandwidth, which is pennies.


The system incurs those fixed costs based on how it anticipates they will pay off. With more money chasing content, the system will find it profitable to green-light more scripts at higher production values. But ultimately everybody’s favorite sound mixer has only two ears.


You can see this in football. In the U.K. satelite TV massively increased the market - no longer were people spending £30 a game to watch at a stadium with a capacity of 30k, they were spending £10 a game to watch on tv with a capacity of millions.

This extra cash poured into the salaries of the footballers in the top division, salary’s doubling every 4 or so years for a sustained period.

The money didn’t trickle down to lower divisions or youth training or on site staff.


We still have to pay for food, shelter, transportation, clothing, and other physical things.

Our inner world is richer, and we consume non-decreasing goods. But everything is tied to something in the physical world, even if it's the hardware and energy running it.


Right, which is why the "100% digital" world was just a thought experiment.... we don't live in that world, but we do live in a world where an increasing percentage of our money is spent on non-exclusionary goods.... that has to have an effect on inflation.


Digital goods prices won't increase due to demand necessarily. Those prices will rise because electricity, rent, and all the other overhead line items increase. Those increases get passed to the customer, who is going to be hit with increased prices for the necessities of life - food, water, housing.

Digital goods do nothing to support human life - food, water, housing. Sure you can buy things online, but you can buy those things at brick and mortar stores. In turn, gas prices will rise to the point where delivery services become unviable.

Inflation touches the entire chain whether the product is digital or otherwise.


Also the marginal cost for Macdonals to serve one more hamburger is also very low. And anything that is automated has very small marginal cost, which is most stuff now a day. If you want to measure inflation, check how much it cost to hire a plumber. And craftsman likes to charge even numbers, so they will jump from $100/hour to $200/hour (rather then from 100 to 110)


The better way to think about this is how much a CPU costs.

To build a modern CPU takes a factory and supply chain that costs multiple billions of dollars. It is a piece of cutting edge technology, sitting at the pinnacle of human engineering achievement. No matter how you rich you get individually, it's not actually possible to through individual effort to buy a CPU which is substantially better then the same thing any consumer can buy.

...so as a result, we sell them to everyone for less then cost of a few weeks groceries.


These organizations would be expected to have some marginal cost increases from running data centers and physical hardware cost increases and they would calculate passing these on to consumers. Or as an excuse to. But I guess that is covered by your example and understanding that their physical costs being non-digital goods. So, fun thought exercise.


Yeah, that is why I intentionally said "practically zero" instead of actually zero, because there are some marginal costs. They are just orders of magnitude less than what they are for physical goods.

But yeah, I am just curious how that math all works out on a macro scale.


This is a good point, and there is also no finite limit which is what typically makes the supply and demand machine go brrrrrrrr.

Netflix pays some % of their monthly subscription fee for servers. The amount really doesn't matter, and they can add the entire planet as subscribers before running out of servers so the supply is infinite for all practical purposes.


Inflation means consumer's purchasing power remains the same, regardless of money supply. Therefore in an inflated economy people won't have the extra money now to subscribe to Netflix as they still are struggling to make ends meet even with the extra money, as everything else has increased in price proportional to his income increase.


> The marginal cost for a new subscriber is practically zero, so there should be no price increase caused by a shortage.

Digital goods aren’t priced at marginal cost. By your logic not only an increase in money supply would have no effect on the price of digital goods, but the price of digital goods should be 0 before and after the increase.


> This isn't the case, however, for digital goods. If there are suddenly 100 million new people who want to buy a Netflix subscription, it isn't like we are going to see the price of a Netflix subscription go up because there isn't enough Netflix to go around.

Good point but you have to think that if there’s lot of demand for Netflix which means Netflix has lot of hit shows. Say their hit percentage is 10% (which is very high). Which means they have to make more and more shows to provide that number of hits to sustain so much demand. Which means more expenses, which puts pressure on them to increase the price of subscription. Which also means actors, story writers etc can charge more (as they are only a finite number of good actors etc)


> Which means they have to make more and more shows to provide that number of hits to sustain so much demand

That's a fixed overhead, because the hit shows costs the same if there is 10 million subscribers, or 100 million subscribers. Netflix doesn't not need to have 10x the number of hit shows to serve 10x the number of subscribers!


I spend fifty times more on real goods each month than digital goods. So while it's fascinating to think about, and it makes quite a lot of sense, it's a rounding error still.


Worth noting that digital goods, almost all that I can think of, are entirely nonessential. They will be the first things cut if a household budget gets tight.


Inflation is not always in relation directly to the price.

For example, if Netflix is able to get 10,000,000 new subscribers. It receives $100,000,000.

It can now spend $50,000,000 on new infrastructure, which will make the price rise.

Netflix can also hire for $50,000,000, which creates jobs that otherwise wouldn't exist. Those new employees, working for that new currency will spend their salaries, rising the prices.


but those $50 million of subscription revenue means there's $50 million less spent on something else. And those $50 million paid to employees are for producing value for netflix (presumably, more than $50 millions worth).

So no, this will not increase inflation. The goods produced matches the money spent, so demand and supply continue to match up.

Inflation would occur if production ceases, but demand continues to remain the same (or higher).


I forgot to specify that the 100 million is entirely printed money.

Whatever the way you see it, that 100 million of buying power should not exist and eventually ends up lifting the prices of normal goods just with the economic activity it generates.


i mean, if you claim that $100 million was printed, then regardless of whether it's netflix or something else, it may increase inflation.

But the thing is, there hasn't been that much money printed by the FEDs or the US gov't. The stimulus cheques are not money printing, but money borrowing - a major difference. Borrowed money needs to be paid back, and so there may be temporary inflation caused by said stimulus, but it's clawed back in the future when the stimulus' effect has worked!


"Borrowed money" is wrong term here. Yes, officially US government is borrowing it, but it is borrowing from FED which does not really have the money.

Imagine this: you need $100k so you come to me, and I write you a check. BUT: you're not allowed to cash that check. Instead you use it as collateral for a $100k bank loan. But the bank is not allowed to cash my check either, it can only put it into safety deposit box. And if you default on your loan bank will put that check on auction. So, time flies an no one ever tries to cash my check. And if they ever tried they would find that I never had that $100k in my account in the first place. I "printed" that money when I wrote the check. That's how "borrowing" between FED and Government works, and that's why it is money printing. FED is bluffing about having all that money, but no one is legally allowed to call their bluff.

Also: no, those loans will never be repaid. With Modern Monetary Theory debt can only increase in nominal value, until US follows the trajectory of Venezuela, becoming a failed state with hyperinflation.


You might want to find out to whom the debt is owed.

The US government owes money to the US government. The trend is an uptrend.

The only thing stopping this stratagem is currency devaluation on the free global market.


The digital version of goods indirectly does cause inflation.

When you have a surge of users and you got more servers in aws, EC2 prices increase for the on-demand instances , aws buys more servers , the more customers it gets , which means more metal excavation , more semiconductors , more minerals , more trucks to move these things , more fuel to move the trucks , it goes on and on.

Also money gets turned into profit , which is then paid to employees , executives , founders.

Who then go on to pay into other items that are non digital (houses , cars , premium cereals lol), which then do cause inflation.

Until all goods are digital , every digital product still indirectly causes inflation using its non digital items , because all of em are transactions between humans.

And all goods can never be digital , because you still need non digital items to run the digital goods on.

So inflation continues.


> The marginal cost for a new subscriber is practically zero, so there should be no price increase caused by a shortage.

This is true in a vacuum. With a bigger customer base a company will need to invest more in support. Also, in the case of Netflix, they need to invest more in sourcing content to cater to the now increasingly varied demands of their customers or else they'll lose them to competition who might also be offering their products at the same rate.


Cantillion effect:

> Cantillion wrote:

> “The river, which runs and winds about in its bed, will not flow with double the speed when the amount of water is doubled.”

> Inflation is not simply an average rise in prices. Prices do not rise proportionally or simultaneously. This results in arbitrary benefit to some who have not created any economic value and detriment to others who have not destroyed anything of economic value by destroying savings for example. This is the Cantillion effect.


Got to add that almost all physical goods are part software these days. In a way, everything becomes software. And fixed costs to create something and put it into the market are huge and tend to grow, but manufacturing unit costs are low and tend to fall.... So this observation applies to most physical goods just as it does to digital ones!


> If there are suddenly 100 million new people who want to buy a Netflix subscription, it isn't like we are going to see the price of a Netflix subscription go up because there isn't enough Netflix to go around.

No, it would go up because they would make more profits with fewer subscribers and a higher margin.


Economies of scale dictate the opposite.

Fixed costs stay the same no matter the subscriber count, this would allow Netflix to lower the price, while maintaining the same profits. In reality, they would likely keep the price the same and increase their margins without charging more.


What's the ratios of fixed to variable cost for Netflix and how many subscribers does it need for the economies of scale to go beyond a few cents?


Businesses charge what customers are willing to pay. If they have more money, they are willing to pay more. Competitiuis the countervailing force, but Netflix has exclusives and serials and network effects (fandoms and friends)


Sure, but that doesn't say anything about inflation.

The standard formula for profit is (units sold * price per unit) - (fixed costs + marginal costs * units sold).... Netflix, like every other company, wants to maximize that profit.

For most non-digital companies, the marginal cost is significant, and follows a u-shaped curve... at first, marginal costs decrease as you sell more units, since you can get intermediate goods for cheaper prices as you buy in bulk. At a certain point, however, the marginal price starts increasing again as you start to hit various bottlenecks and intermediate goods start becoming more expensive as you consume all the easy to produce supply. In other words, you can't scale linearly.

Digital goods have a much flatter uptick on that marginal cost graph, and I am very curious what that means at the macro economic level.


It means that in your formula, the (units sold * price per unit) term dominates the profit formula. So, the price is mostly determined by the consumer's willing to pay.


And by your competitors willingness to take a lower per-unit profit in exchange for more sales.


If you permit me to be terribly econ 101 about this, if netflix had a perfectly elastic supply curve (because the marginal cost of an extra netflix subscriber is fixed and doesn't increase with high numbers of subscribers) then an increase in demand would lead to an increase in quantity supplied but not an increase in price.


If you're going to be "terribly econ 101", note that Netflix is not in a perfectly competitive market and almost certainly encounters a downward sloping demand curve... and that its maximum profit point is not going to be at the point where the most units are supplied and may indeed shift as the demand curve shifts.


Many people speak with great confidence about inflation, the money supply and "econ 101". Most of those with the greatest confidence in their own knowledge are not familiar with the fundamental equation of exchange, MV=PQ.

Here, M is the money supply, V is the velocity of money, P is the price level, and Q is the real quantity of goods and services.

It's easy to see that if M increases and Q increases the same amount, then if V is constant, P will also hold constant.

Or if Q is constant and a decrease in V offsets the increase in M, P will again hold constant.

Increasing the money supply by printing money does not automatically cause prices to increase, because there are other variables in the equation.


Thanks, and I understand that... but I feel like you don't understand the context of my reply.

Mr. Beer above stated that an increase in demand for Netflix would not increase the market price of Netflix. But because Netflix is a quasi-monopoly, they would be likely to increase prices to find the new equilibrium maximum profit point.

That is, he made a microeconomic argument which was invalid-- arguing that because Netflix's marginal cost is low the price elasticity of the product they supply must necessarily be low-- but this is not necessarily true. Netflix has some degree of pricing power.

I made no macroeconomic argument, and.. while I agree with your macroeconomic assertions... increasing the supply of money still increases inflationary pressures. If you're facing massive deflationary headwinds because of slowing velocity of money, sure, it may not be enough to cause inflation or even entirely prevent deflation, but it still is net inflationary.


My reading of GPs comment was more about the cause and effect of any corresponding price increase, not whether Netflix would actually increase their price in this scenario.

For a physical good, if you have a surge in demand as more people can afford you're product you start running into supply issues. This will likely result directly in a price increase as you can't increase profit by just selling more when you don't have anymore to sell.

In the example of netflix, the supply issue almost completly disappears. They don't HAVE to increase the price to increase their profit. Yes Netflix may decide that with the flood of new customers they can put the price up and maintain or increase their profit.

The point I think that is being made is that their isn't a physical supply issue in effect "forcing" them to increase the price. They are doing it because they want to, not because they need to.


Sure, there's no "force" to increase the price. But finance and marketing wonks read books like "Pricing and Revenue Optimization", and seeing a big upswell in demand are tempted to do the math again and see what the new maximum profit point is if they have pricing power.


If the marginal cost of a new subscriber is 0 then increasing their costs should only decrease revenue.


>But because Netflix is a quasi-monopoly, they would be likely to increase prices to find the new equilibrium maximum profit point.

That's the thing though, if they have a perfectly horizontal supply curve, then the maximum profit point would move horizontally but not vertically, and price would not increase. This sort of analysis is surely too simplistic, but that's the model.


Yeah, but that is redirecting the focus away from the major problem that we face in practice, which is that M doubles and the money gets given to people who didn't earn it. To add insult to injury, they are also usually wealthy and taking extreme risks that destroy value.

It doesn't matter if prices double, remain constant or halve. It matters is people can afford more, the same or less stuff. At the moment, the rapid pace of technological advancement means most people should be able to afford much more and they can't because of the incessant money creation being done by people in charge of the system.


It isn't "given" though, it's loaned, or used to purchase an asset (generally a government bond which the government is obliged to pay the holder coupon payments on in future if the holder doesn't sell)


Slightly off topic, but talking about absolute price levels does not make much sense to me. Let's say V goes down, because consumers have discovered the value of thrift and are storing cash under their mattresses. Prices then should come down, to reflect the new economic reality, send signals to decrease production etc.

Then, if money is unexpectedly debased, absolute price levels may not change, but they would go up relative to where they should have been. Contracts still end up being distorted. People with cash under their mattresses still end up with their holdings devalued.

I wonder, if one had the task of building an economic system from scratch, if they would come up with the current system or with something else. The current system seems like a mess of patches upon patches, many heuristics, and is not very philosophically sound.


Actually no, even if they face a downward sloping demand curve (which everyone does but I know what you mean) if their supply curve is perfectly horizontal (infinite elasticity) then an increase in demand would still hold price steady and see only an increase in quantity supplied.


"If they have more money, they are willing to pay more."

That's like totally wrong. Goods provide some value to the customer, and that determines what they are willing to pay for it. If I wake up a millionaire tomorrow, that doesn't mean I am suddenly willing to spend $500 the same haircut that was $50 yesterday.


> Goods provide some value to the customer, and that determines what they are willing to pay for it

Very often customers will pay twice as much or more for the same or equivalent good that delivers the same value, you're local grocery store will show some examples. People will often pay 5x more for brand name items over white label ones. Stores will very regularly discount some items to take advantage of price discrimination, people not paying attention to specials are effectively paying twice as much for the exact same items. If you go to a different grocery store in a wealthier/poorer area you'll see the exact same items at different price points. These are all ways companies will maximize profits because people willingly pay more for the same items.


Once you are a millionaire for a few years your brain requires itself around value of time.

I’m not a millionaire but I’m certainly well off, and I just happily overpaid for lumber for a project. At prices that three years ago would have made me cancel the project. Not because I couldn’t afford it but because the perceived value wasn’t worth the price.


American has too many monopolies though, and I believe price collision in many industries is alive, and doing well, but has gotten harder to prove.

My mom wants to nix Comcast, and it's not that easy.


They don't price things based on cost, they price them based on what you are willing to pay.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: