great post albert. i think of the economics of abundance as the economics of personalization -- i.e. the economics of a world in which we can increasingly create our own world.
albert
· 8 months ago
Personalization is hugely important and having a single price for a song (the current model) is a big obstacle for it.
Salman FF
· 8 months ago
Is "Price Discrimination" scalable? It's not just "because digital goods are so easy to redistribute [that] they don’t meet the traditional criteria which would allow for price discrimination". It's also that the information that points to (or links to) the lowest price option is so easily (and freely) available and accessible. Price discrimination is already happening on the web in various forms, but how can you make it scale reliably? Any examples you can think of? BTW - It was a great post. Very deep. Looking forward to the sequels.
Q dub
· 8 months ago
The thing about digital goods is that today, they are non-substitutes of each other, so every virtual item can be its own monopoly. What I mean is, even Mafia Wars godfather points is sufficiently distinct from Mob Wars (you can't just pack-up your 100-person mafia and switch games) that you shouldn't expect prices to hit marginal cost ($0). Same could be said for music and movies.
Now onto price discrimination: Name-your-price is actually the hardest kind of price discrimination (first degree)...but you don't have to strive for that degree of perfection. I think a lot of the action we'll see is in second degree--playing with volumes and service tiers. Like a song? Buy the song. Love it? Buy the song and a few live versions. Can't live without it? Buy the song, live performances, and a couple remixes. Like this game? Play for free. Love it? buy a few virtual goods and enhance your experience even more. That doesn't seem so un-scalable to me!
albert
· 8 months ago
Completely agree that we will see many forms of second degree price discrimination. You are also right that points in games cannot be substituted for each other. They can also be kept artificially scarce by the folks who control the game, because they are part of a closed system. It is therefore correct that one should *not* expect their price to go to zero. But "same could be said for music and movies" does not follow because they are not part of a closed system in which artificial scarcity can be maintained.
Salman FF
· 8 months ago
I am not sure you can equate "freemium" type models (which have proven quite scalable and successful from way before the internet) with xth order price discrimination, which is about segmenting the market for the same product.
Dmitriy
· 8 months ago
One of the all-time greatest blog posts - thank you very much!
I am not personally convinced that customer-controlled price discrimination (AKA name-your-own-price) in digital goods is the ultimate solution though. It may work in certain cases or even in many cases, but it needs a lot of relatively-honest customers (thousands? millions?) Not every market is that big (think your local news outlet). Additionally, many people overlook the effect of novelty on paying decisions in this scheme. Yes, Radiohead experiment was great, but I am almost 100% sure many people will not pay the same amount the next time they get such an opportunity. I suspect that when presented a name-your-own-price option for the 100th time, a lot of people will just ignore it. Do you know of any studies focusing on how people react to "name your price option" over time? I will be very surprised if severe novelty effect did not exist here.
Finally, I think name-your-own price distorts (not completely eliminates) signaling function of prices, because supplier doesn't contribute anything to the pricing decisions, and hence they can't tweak one thing and observe effects, in order to take this into account next time.
Instead, I have always been thinking that better way to price digital goods (ones with 0 marginal cost) is via subscription based on time spent with the service, not based on units of consumption (stories read, songs listened to) and not with an upfront pricetag (for example, $5/month).
For example, NYTimes may charge $0.001 per minute spent on their site. (I know such technology does not exist for this yet). The key is not to bill based on consumption of units (songs, articles) but on time spent engaging with the service.
IMHO, this has a higher chance of working out long term, but I may be wrong :)
P.S. I am a software systems engineer and economics is my hobby, so please feel free to discard this all as just a pile of nonsense...
albert
· 8 months ago
Thanks - agree that this may only be one component. Did a follow up post two days later that addresses some of the issues you raised.
perhaps, you can price discriminate on past behavior (using past purchases as indicative of future valuations for similar goods) and for first experience goods charge a standard price/or time spent (as mentioned in the comment directly above).
albert
· 8 months ago
That's definitely useful information that can be used to support certain types of 'second degree' price discrimination.
All Comments
It's not just "because digital goods are so easy to redistribute [that] they don’t meet the traditional criteria which would allow for price discrimination". It's also that the information that points to (or links to) the lowest price option is so easily (and freely) available and accessible. Price discrimination is already happening on the web in various forms, but how can you make it scale reliably? Any examples you can think of?
BTW - It was a great post. Very deep. Looking forward to the sequels.
Now onto price discrimination: Name-your-price is actually the hardest kind of price discrimination (first degree)...but you don't have to strive for that degree of perfection. I think a lot of the action we'll see is in second degree--playing with volumes and service tiers. Like a song? Buy the song. Love it? Buy the song and a few live versions. Can't live without it? Buy the song, live performances, and a couple remixes. Like this game? Play for free. Love it? buy a few virtual goods and enhance your experience even more. That doesn't seem so un-scalable to me!
I am not personally convinced that customer-controlled price discrimination (AKA name-your-own-price) in digital goods is the ultimate solution though. It may work in certain cases or even in many cases, but it needs a lot of relatively-honest customers (thousands? millions?) Not every market is that big (think your local news outlet). Additionally, many people overlook the effect of novelty on paying decisions in this scheme. Yes, Radiohead experiment was great, but I am almost 100% sure many people will not pay the same amount the next time they get such an opportunity. I suspect that when presented a name-your-own-price option for the 100th time, a lot of people will just ignore it. Do you know of any studies focusing on how people react to "name your price option" over time? I will be very surprised if severe novelty effect did not exist here.
Finally, I think name-your-own price distorts (not completely eliminates) signaling function of prices, because supplier doesn't contribute anything to the pricing decisions, and hence they can't tweak one thing and observe effects, in order to take this into account next time.
Instead, I have always been thinking that better way to price digital goods (ones with 0 marginal cost) is via subscription based on time spent with the service, not based on units of consumption (stories read, songs listened to) and not with an upfront pricetag (for example, $5/month).
For example, NYTimes may charge $0.001 per minute spent on their site. (I know such technology does not exist for this yet). The key is not to bill based on consumption of units (songs, articles) but on time spent engaging with the service.
IMHO, this has a higher chance of working out long term, but I may be wrong :)
P.S. I am a software systems engineer and economics is my hobby, so please feel free to discard this all as just a pile of nonsense...