4 ms·

Could you clarify this a bit? It sounds like an auction would just be a more complicated way of raising prices, since the tickets would still go to those custo

Could you clarify this a bit? It sounds like an auction would just be a more complicated way of raising prices, since the tickets would still go to those customers who are able to bid the most.

qsort 6y ago
You can run auctions in a bunch of different ways; in particular Vickrey–Clarke–Groves has rules such that the optimal strategy is for each participant to bid their true evaluation. Of course the prices are going to increase. If the supply is the same and the demand increases, the price increases. There is nothing anybody can do about that.
- AndrewDucker 6y ago
  FTA: "Yes, we’d get to line our pockets with money for days, but we like to think of PAX as the show for the every-gamer, and if we didn’t keep badges affordable, it’s not fair to those that can only afford our current price."
  - qsort 6y ago
    Scalpers are already effectively increasing the price. If the price of the event is $100 and I value it at $300, I'm going to buy from a scalper at $200. It would be more fair to "distribute" the price increase among all participants by using an auction method that incentivizes people to bid their true evaluations. I don't want to be unfair to the guy, he seems to be understanding the problem and it's not an easy decision, but that sentence is a bit of a platitude.
    - AndrewDucker 6y ago
      As they say, it's only affecting about 1% of their tickets. So 99% of people aren't having their prices inflated.
      - qsort 6y ago
        If that is the case, then it follows that the expected price increase would be approximately the same in a Vickrey auction (barring extreme cases where the bids have sharp discontinuities).
        ivanbakel 6y ago
        How does that follow? The PAX organisers are "artificially" keeping down the price of tickets by controlling supply and fighting scalpers. If instead the auction allowed potential attendees to pay more for their tickets, then those tickets would end up going to attendees that pay far above the original price - as is already true for scalped tickets. The percentage of tickets which are scalped is not a direct function of the demand on tickets.
        qsort 6y ago
        Because if a ticket is traded at $X, it means that there exists someone who believes it is worth $X or more. PAX is effectively offloading the additional cost to the excluded audience, by either denying a ticket they would be willing to buy, or by making them pay more than they would have. If the fraction of the scalped tickets is low, then it means that the fraction of people who are willing to buy a ticket at a price much higher than the going price can't be that large. If tickets were sold at $1, you'd certainly see A LOT more scalping going on, wouldn't you? This would in turn imply that the global outcome wouldn't be that different. People who really want a ticket get one either way, and people who kinda want a ticket benefit from the more fair allocation system.
        ivanbakel 6y ago
        >If tickets were sold at $1, you'd certainly see A LOT more scalping going on, wouldn't you? Not necessarily - you're discounting the effectiveness of anti-scalping measures. (And moreover, scalpers are always incentivised to scalp as many tickets as possible, because they are then put in the situation of controlling the market price with reduced supply, so that they can jack up prices). If in fact anti-scalping works effectively, then the market price of the ticket in a fair auction can be completely independent of the original price it was sold at. If demand is high enough, you can even have people who can only buy tickets because of the fixed price - if even 20% of customers are willing and able to pay far above the retail price for a ticket, they would in an auction snap up all the tickets, because PAX is oversubscribed by 5-to-1.
- 3np 6y ago
  Myself I was thinking it’d make sense with some variation of Dutch auction in rounds, where all participants in a round get the same price. What would you say the upside of VCG is? Dutch seems simpler and fair to me. https://en.wikipedia.org/wiki/Dutch_auction https://en.wikipedia.org/wiki/Dutch_auction
  - qsort 6y ago
    A Dutch auction is more or less strategically equivalent to a first-price, sealed-bid auction. The advantages of VCG over that would be: - it's easier to implement. Dutch is admittedly easier to understand, but with VCG your average buyer would only need to fill a box in a form. - there is provable dominance of truthful bidding There are also downsides, admittedly: - it doesn't maximize seller revenues - it doesn't work if all bidders collude (unlikely in this scenario) - a dishonest seller could use shill/fake bids to increase the going price
    - philihp 6y ago
      Call me chaotic neutral, but an auction for a Penny Arcade convention should really be a penny auction.
- ginko 6y ago
  >Of course the prices are going to increase. If the supply is the same and the demand increases, the price increases. There is nothing anybody can do about that. The profit over the targeted ticket price could be returned uniformly to ticket holders. I guess this would be hard to do in practice and may complicate bidding strategies but it would avoid price increases.
  - qsort 6y ago
    Doesn't that defeat the purpose of the auction in the first place? That would simply become "the ticket goes to whoever writes the largest number on this piece of paper", right? I believe to reach your intended goal (i.e. the seller doesn't want to profit from the auction), you would have to redistribute the price increase among all participants. A way to do this would be as follows (needless to say, completely unpractical fantasy land experiment, I'm only doing this for the sake of argument): - Have people freely register. Let's say you have M tickets, N people registered, the interesting case is when N >> M. - Assign to each registered participant M/N of a ticket (assume tickets are infinitely divisible), and collect from them M/N * P where P is the intended price of a single ticket. - Have people trade fractions of tickets. The seller collects exactly M * P, and under usual assumptions the price of a ticket would converge to its 'true' value.
    - ginko 6y ago
      > Doesn't that defeat the purpose of the auction in the first place? That would simply become "the ticket goes to whoever writes the largest number on this piece of paper", right? People who are bidding more will still pay more. The main weirdness introduced by that approach is that low bidders that win a ticket may get more money back than they paid.

Could you clarify this a bit? It sounds like an auction would just be a more complicated way of raising prices, since the tickets would still go to those customers who are able to bid the most.

qsort 6y ago

You can run auctions in a bunch of different ways; in particular Vickrey–Clarke–Groves has rules such that the optimal strategy is for each participant to bid their true evaluation. Of course the prices are going to increase. If the supply is the same and the demand increases, the price increases. There is nothing anybody can do about that.

AndrewDucker 6y ago
FTA: "Yes, we’d get to line our pockets with money for days, but we like to think of PAX as the show for the every-gamer, and if we didn’t keep badges affordable, it’s not fair to those that can only afford our current price."
- qsort 6y ago
  Scalpers are already effectively increasing the price. If the price of the event is $100 and I value it at $300, I'm going to buy from a scalper at $200. It would be more fair to "distribute" the price increase among all participants by using an auction method that incentivizes people to bid their true evaluations. I don't want to be unfair to the guy, he seems to be understanding the problem and it's not an easy decision, but that sentence is a bit of a platitude.
  - AndrewDucker 6y ago
    As they say, it's only affecting about 1% of their tickets. So 99% of people aren't having their prices inflated.
    - qsort 6y ago
      If that is the case, then it follows that the expected price increase would be approximately the same in a Vickrey auction (barring extreme cases where the bids have sharp discontinuities).
      - ivanbakel 6y ago
        How does that follow? The PAX organisers are "artificially" keeping down the price of tickets by controlling supply and fighting scalpers. If instead the auction allowed potential attendees to pay more for their tickets, then those tickets would end up going to attendees that pay far above the original price - as is already true for scalped tickets. The percentage of tickets which are scalped is not a direct function of the demand on tickets.
        qsort 6y ago
        Because if a ticket is traded at $X, it means that there exists someone who believes it is worth $X or more. PAX is effectively offloading the additional cost to the excluded audience, by either denying a ticket they would be willing to buy, or by making them pay more than they would have. If the fraction of the scalped tickets is low, then it means that the fraction of people who are willing to buy a ticket at a price much higher than the going price can't be that large. If tickets were sold at $1, you'd certainly see A LOT more scalping going on, wouldn't you? This would in turn imply that the global outcome wouldn't be that different. People who really want a ticket get one either way, and people who kinda want a ticket benefit from the more fair allocation system.
        ivanbakel 6y ago
        >If tickets were sold at $1, you'd certainly see A LOT more scalping going on, wouldn't you? Not necessarily - you're discounting the effectiveness of anti-scalping measures. (And moreover, scalpers are always incentivised to scalp as many tickets as possible, because they are then put in the situation of controlling the market price with reduced supply, so that they can jack up prices). If in fact anti-scalping works effectively, then the market price of the ticket in a fair auction can be completely independent of the original price it was sold at. If demand is high enough, you can even have people who can only buy tickets because of the fixed price - if even 20% of customers are willing and able to pay far above the retail price for a ticket, they would in an auction snap up all the tickets, because PAX is oversubscribed by 5-to-1.
3np 6y ago
Myself I was thinking it’d make sense with some variation of Dutch auction in rounds, where all participants in a round get the same price. What would you say the upside of VCG is? Dutch seems simpler and fair to me. https://en.wikipedia.org/wiki/Dutch_auction https://en.wikipedia.org/wiki/Dutch_auction
- qsort 6y ago
  A Dutch auction is more or less strategically equivalent to a first-price, sealed-bid auction. The advantages of VCG over that would be: - it's easier to implement. Dutch is admittedly easier to understand, but with VCG your average buyer would only need to fill a box in a form. - there is provable dominance of truthful bidding There are also downsides, admittedly: - it doesn't maximize seller revenues - it doesn't work if all bidders collude (unlikely in this scenario) - a dishonest seller could use shill/fake bids to increase the going price
  - philihp 6y ago
    Call me chaotic neutral, but an auction for a Penny Arcade convention should really be a penny auction.
ginko 6y ago
>Of course the prices are going to increase. If the supply is the same and the demand increases, the price increases. There is nothing anybody can do about that. The profit over the targeted ticket price could be returned uniformly to ticket holders. I guess this would be hard to do in practice and may complicate bidding strategies but it would avoid price increases.
- qsort 6y ago
  Doesn't that defeat the purpose of the auction in the first place? That would simply become "the ticket goes to whoever writes the largest number on this piece of paper", right? I believe to reach your intended goal (i.e. the seller doesn't want to profit from the auction), you would have to redistribute the price increase among all participants. A way to do this would be as follows (needless to say, completely unpractical fantasy land experiment, I'm only doing this for the sake of argument): - Have people freely register. Let's say you have M tickets, N people registered, the interesting case is when N >> M. - Assign to each registered participant M/N of a ticket (assume tickets are infinitely divisible), and collect from them M/N * P where P is the intended price of a single ticket. - Have people trade fractions of tickets. The seller collects exactly M * P, and under usual assumptions the price of a ticket would converge to its 'true' value.
  - ginko 6y ago
    > Doesn't that defeat the purpose of the auction in the first place? That would simply become "the ticket goes to whoever writes the largest number on this piece of paper", right? People who are bidding more will still pay more. The main weirdness introduced by that approach is that low bidders that win a ticket may get more money back than they paid.