Comment by klagermkii | Hacker News Reader

Comment by klagermkii | Hacker News Reader

I don't agree with the idea that giving companies better and better tools to achieve price discrimination is the win-win that the final quote seems to make it out to be. Yes, I can see how it could allow a company to offer goods to people who otherwise would not be able to access their services, but this comes at the cost of potentially complicated usage restrictions, making it easier for incumbents to squash new competitors, and pushing for more and more of a legal framework to support this kind of business model at the expense of the end user. I can think of examples where the products are barely (if at all) worse and see how heavily it swings things in favour of the powerful incumbent:

Trademark restrictions to prevent importing what could be effectively the same textbook or wrist watch.

DVD region locking where you can sell practically an identical product but charge vastly different rates in different areas, and it's illegal to work around.

License agreements on software not to restrict specific functionality, but rather the usage scenarios.

The example in the article of the slower LaserPrinter example morphs from being a physical restriction to modern printer cartridges that block ink refills via DRM.

Comcast's battle with Google Fibre where the price will heavily drop in just the areas where Google Fibre operates. Makes it much harder for Google Fibre to make any gains, but the vast majority of Comcast customers see no decrease in their bills.

Even the recent EpiPen drama, where they want to be able to milk as much as possible out of the pockets of insurance companies, but end users don't want to play the real price, so they introduce the co-pay discount to try and offset that cost. It makes the "rich" organisation pay, but reduces the amount paid by an individual.

The devils advocate is something like:

If a game cost say 50$ they sell approximately zero of them in India. Dropping that to 5$ does not mean they sell 10x as many worldwide thus company keeps the higher price. If they can sell them for 5$ just in India they may sell 10x as many in India which is good for people in India and the Company. But, people in US are never getting 5$ disks either way. Making this a net win.

The counter argument is price discrimination often operates inside a country. Coupons are the classic example, but clearly Coupons are a net economic loss.

Which IMO suggests we should have limits on the kind of acceptable price discrimination.

> clearly Coupons are a net economic loss

I'm not so sure about that. Yes, the process of printing, distributing, and using coupons is a deadweight loss to society, but the ability to price discriminate can create surplus.

If the captured producer surplus were less than the cost to print and distribute coupons, then manufacturers would stop doing it.

This is exactly analogous to your example with India as well. There is a deadweight loss associated with administering and enforcing regional trade barriers so that you can't just go buy the $5 Indian version of a good.

This assumes it's not a zero sum market. If a company can extract more money than it costs them to print Coupons they will do so. But, coupons may simply allow them to raise average prices without selling more or higher quality goods.

Basically, if you sell the 100 widgets for 10$ that's 1,000$. But if you use coupons to sell the good for 30 widgets for 9$ and 70 widgets for 12$ at the cost of 20$ that's 1110$ - 20 = 1090$. It's the same number of sales in the second place, but your profit is higher.

Thus, just because it's more profitable does not mean it's a net economic gain. Monopolies are great when they are your Monopoly.

> This assumes it's not a zero sum market

Economics in general is not a zero sum market. People trade because they see benefits to doing so. I buy something because the value I derive is more than the cost -- that's why economies and markets create value. We call that value 'surplus' (https://en.wikipedia.org/wiki/Economic_surplus)

> just because it's more profitable does not mean it's a net economic gain

In the specific example you made, yes, the producer is just capturing more of the surplus, more hasn't been created. In general, price discrimination shifts surplus from consumers to producers.

However, it also shifts surplus from richer buyers to poorer ones. So coupons are a benefit to the poor, and airline price discrimination benefits you if you buy 3rd class tickets. Unless you are very rich, price discrimination is generally a monetary benefit to you.

Note that 3rd-degree price discrimination can increase or decrease total welfare depending on the specifics:

http://www.econ.ucla.edu/hopen/econ171/monopoly2.ppt (see slide 4)

> Unless you are very rich, price discrimination is generally a monetary benefit to you.

You're not accounting for the deadweight loss from the crippled product.

Suppose if there were no coupons the product would cost $20 to everyone, vs. when coupons are used it costs $15 with coupon or $30 without. But using the coupon requires an hour worth of labor on the part of the customer. So the rich customer pays the $30 and the poor customer pays the $15 + $10 worth of labor. Compared to the case with no price discrimination, the rich customer pays $10 more and the poor customer pays $5 more. Neither customer is ahead, only the seller is.

Likewise with crippled products, if the full version of the product was available for $20 to everyone then the poor customer may get $25 worth of value from it and have a $5 surplus. Making the full version $30 and the crippled version $15 means the poor customer won't buy the full version, but the crippled version may only provide $16 worth of value, so now the poor customer's surplus is down to $1 from $5.

You're also assuming price discrimination costs the rich more than the poor. But volume licensing and bulk discounts are a form of price discrimination.

In the limit where the seller can engage in perfect price discrimination, the seller captures ~100% of the consumer surplus from all consumers, which is obviously worse for every consumer than any case where the consumer gets to keep any part of the surplus.

Sure, you can always find parameters where one party or another is hurt, because the parameter space is very large.

For example, without the excess profits provided by price discrimination, a company may decide a market is simply too small to serve, in which case all surplus is lost.

Also, the baseline for comparing a crippled product is not necessarily the same but uncrippled product. If a manufacturer knows crippling is unavailable, they may just develop the reduced feature set entirely, in which case rich consumers are hurt.

Microeconomic theory can be subtle, and whether something is good or bad depends very much in the specific case.

> Also, the baseline for comparing a crippled product is not necessarily the same but uncrippled product.

Conversely, the baseline for the lower priced product is not necessarily lower than the existing product. If the full version is currently $50, the profit-maximizing strategy could be to sell a crippled version for $50 and the full version for $500, leading to more profit for the seller but less surplus overall.

> Microeconomic theory can be subtle, and whether something is good or bad depends very much in the specific case.

That's what I'm saying. Your argument was that price discrimination generally helps the poor. But it only does when the specific method of price discrimination happens to do that entirely by accident -- if it happens it's because the seller set the price of the lower cost version at less than they actually needed to, or didn't cripple it as much as they could have which would have increased sales of the more expensive version.

The one thing price discrimination is designed to do is convert consumer surplus into producer surplus. Which in general is not good for the consumer.

I think we're confounding a few different facts here:

* Effect of 3rd-degree price discrimination on total surplus can be positive or negative, and that was the point I was trying to make

* We both agree that price discrimination transfers surplus from consumers as a whole towards the producer

* However, it's easy to show that at least in the simple case of perfect 3rd-degree price discrimination, the 'poor' consumer is either indifferent or is made better off when the producer can price discriminate

Sketch of a proof: In a 3rd-degree price discriminated market, let P1 and P2 be the reserve prices of 'poor' and 'rich' consumers, respectively. Now let's get rid of price discrimination -- what's the new profit-maximizing price for the producer?

It's easy to show that it cannot be less than P1 or more than P2, so the new price P* lies in [P1, P2].

In that case, all the 'poor' consumers that were paying P1 must now pay P* or not transact at all, so they are all now worse off.

Note that you have been emphasizing the transfer of wealth from consumer to producer, but classical microeconomics focuses most of its attention towards total surplus, which includes producer profit. This is because you can always find ways to divvy up surplus once it's been created to make things more fair (e.g. by taxing the producer or by distributing equity around), but you can never recover deadweight losses.

> In that case, all the 'poor' consumers that were paying P1 must now pay P* or not transact at all, so they are all now worse off.

Except that P1 and P2 cannot be different unless what you get for them is different, or else the rich customers would also choose P1. So the poor who pay P1 pay less money but in exchange for an inferior product, which is not guaranteed to yield a net positive outcome.

You're right, but the irony here is that perfect price discrimination (where the producer can costlessly and effortlessly force rich customers to pay P2) is the best possible situation for the poor customers.

This implies that relatively cheap ways of borking a product (make one chip, disable some features) can be better for low-end consumers than more expensive methods (such as actually running two production lines), even though it seems less fair.

> You're right, but the irony here is that perfect price discrimination (where the producer can costlessly and effortlessly force rich customers to pay P2) is the best possible situation for the poor customers.

The poor consumer is agnostic to how much of the surplus is captured by the rich consumer vs. the producer. It's possible for that transaction to affect the poor consumer, e.g. if the producer gets it and spends it on R&D or the rich consumer gets it and spends it on some other public good, either of which could benefit the poor consumer. But either of them could also use it for something adverse to the poor consumer's interests, or just put the money in their pocket and walk away.

And just because the producer is extracting as much as possible from the rich consumer doesn't mean it isn't also extracting as much as possible from the poor consumer -- doing that is the definition of perfect price discrimination.

If manufacturers knew that the captured producer surplus were less than the cost to print and distribute coupons, then manufacturers would stop doing it.

Nowadays US customers will easily learn the same game sells in India for $5, and get angry. So that has to be taken into account as well.

As a person from a Nordic country I used to buy games from Russia and activate them on Steam. Nowadays I play fewer games so I can afford to pay the full price of new titles, but in the past, I used to save 2/3 of the game's price by logging through a VPN from Russia and buying the game that way.

Steam has said this kind of activity to be punishable and sometimes people get their account locked. However when you can save 40 euros on a game title by spending ten extra minutes on a VPN connection then it kind of makes it worth it. Gambling this way is also thrilling.

It's a hard issue to address and ultimately there is no way Steam can effectively block out people from doing it. I could appeal any ban by saying that I visited Russia in person, bought the physical game there and activated it on a cafe or through $VPN_PROVIDER.

Pricing products differently in different countries is all about charging whatever people will pay for something, in order to maximize profit.

Australia's minimum wage is ~$16/hr, so Adobe products, Apple, even online downloads cost a lot more than the US, because people in Australia have more money to spend.

When the citizens of a country have a lot of spending money, international companies charge them more than in countries where the citizens don't have a lot of money.

Viewed this way, it's really interesting to note that "stuff" (cars, laptops, phones, beer, smokes) is cheaper in America than any other developed country in the world. This is an interesting way to come to the conclusion that the average American citizen has less spending money than the average developed country citizen

Good point, but you're leaving out the effects of taxes and regulation on prices.

Cigarettes are jaw-droppingly expensive in Australia because the government wants as few people smoking as possible. The icky labels on packages don't make it very attractive either.

Whereas in America, cigarettes can range in price from annoyingly pricey to cheap as dirt, especially if you buy in southern states or Native American reservations (another price distortion).

>American citizen has less spending money than the average developed country citizen

You seem to be confusing real and nominal prices. Australian's merely have more dollars worth less, in goods, each. That's not necessarily a "good" thing. In fact, it's likely a negative.

The AUS minimum wage is higher, but not the per-capita mean. Stuff is cheaper in the US primarily because the market is larger, so fixed costs can be spread over more units sold. And larger markets also mean more competitition, which means companies can't as easily price discriminate. Government regulation, and the relative isolation of Australia probably contribute as well.

> but not the per-capita mean

The interesting thing about the per capita mean in the US is that it's completely meaningless. So few have so much, that it doesn't tell us anything about how much money regular folks on the street actually have to spend on stuff.

I don't think its completely meaningless any more (actually significantly less) than using a country's minimum wage to make inferences.

I'd buy that if we were talking about the median, but the mean? It really is completely meaningless on its own.

I can tell you never lived in a poor country.

Often times it's the reverse: international goods are same or significantly higher than USA or first world prices: the west embraced free trade but the rest of the world not so much....

A few things like McDonalds will be lower by 30% but many chain stores will go the opposite direction: luxury pricing premium...reasons vary from local tariff(especially for cars or electronics) or just brand snobbery(Nike is expensive in china for example) or protective laws(Korean electronics and cars are more expensive in Korea than USA, tariffs keep out cheaper foreign competitors).

Often times, poor countries have crazy bad wealth distribution ratios so overcharging to enhance prestige is fairly normal. Going low on price is a bad strategy since 90% of the population can't even afford your goods but the top 5-10% can spend like kings.

Australia getting screwed on pricing is likely due to some goofball middle man system or taxing/tarriff scheme combined with island isolation surcharge(hawaii suffers this too).

I wouldn't be surprised if their is some foreign software tax.

> I can tell you never lived in a poor country.

On the contrary, I spent 2 years in Central and South America (almost lived in Argentina) and I'm now making my way down the West Coast of Africa.

That aside, I wasn't really talking about "poor" countries, I was talking about the pricing of stuff like cars, TVs, laptops among developed countries.

People from Europe/Australia are continually staggered when I tell them how little I paid for a 4x4, or an Apple Laptop in the USA.

Good points and examples here, since the article doesn't address these common types of scenarios at all. Some are less problematic, but others do present significant issues. Here's my take:

> Trademark restrictions to prevent importing what could be effectively the same textbook or wrist watch.

The case law on this is mixed. the Costco-Omega case [1] was a 4-4 decision that bears no precedential weight, and is from the 9th Circuit (this means it could be cited but is not binding on lower courts). It was decided in favor of Omega, which sought to prevent importation by Costco, notwithstanding the First Sale Doctrine (which generally says that after you buy a copyrighted article, you can do what you want with it). If this were a binding case from a high court, this would be bad. It is much less unfortunate because it is not binding, and because a subsequent case from a higher court went differently.

The textbook case (Kirtsaeng-Wiley) went the other way, and allowed importation [2]. This was a US Supreme Court case and was decided by a majority. So on balance, the law on this seems to impair the effectiveness of price discrimination, which is good for the consumer. This case did not overrule the Omega case, which was factually distinguishable, so this doesn't mean that there aren't still potential issues lurking.

> DVD region locking where you can sell practically an identical product but charge vastly different rates in different areas, and it's illegal to work around.

DVD region locking, and especially the DMCA overlay, is definitely problematic. This applies equally to printer cartridge refills. The only potential benefit for consumers is that printer companies (and their analogs in other industries) can make printers less expensive, so that more people can afford them. If they couldn't lock you in to their ink cartridges, then they would have to charge more up-front. The counterargument to this is that many people don't understand the cost structure and end up locked in. Some printers even come with part-full cartridges, which is particularly deceptive.

> License agreements on software not to restrict specific functionality, but rather the usage scenarios.

This sort of thing can be done in "bad" ways, but it can also be done to offer academic discounts, discounts for small startups/businesses (think Slack's model). I'm not sure how one would draw the line between beneficial and nefarious applications of this model.

> Comcast's battle with Google Fibre where the price will heavily drop in just the areas where Google Fibre operates. Makes it much harder for Google Fibre to make any gains, but the vast majority of Comcast customers see no decrease in their bills.

Completely agree—this relates to there being a monopoly in many areas, which throws everything out of whack. Consumers definitely lose here, on the whole. (This is also why I wrote this article [3] about one way to hedge this problem.)

> Even the recent EpiPen drama...

This is mostly a problem because of the way health insurance presents a principal-agent problem. The person making the decision (insured) is not the person paying for them (insurance company). This definitely creates problems. It is arguably a good thing that medical device/drug companies can offer discounts to lower-income folks. On the whole, not sure how to weigh the pros/cons in the medical industry generally.

1 https://en.wikipedia.org/wiki/Omega_S.A._v._Costco_Wholesale...

2 https://en.wikipedia.org/wiki/Kirtsaeng_v._John_Wiley_%26_So...

3 https://medium.com/@nicklum/how-my-hatred-of-comcast-drove-m...

Reply on news.ycombinator.com