Comments on: Network Effects = Network Externalities http://www.midasoracle.org/2007/10/23/network-effects-network-externalities/ Prediction Markets, etc. Thu, 02 Feb 2012 23:13:18 +0000 hourly 1 http://wordpress.org/?v=3.3.1 By: Niall O'Connor http://www.midasoracle.org/2007/10/23/network-effects-network-externalities/#comment-16431 Niall O'Connor Sun, 28 Oct 2007 21:14:33 +0000 http://www.midasoracle.org/2007/10/23/network-effects-network-externalities/#comment-16431 Having compared two hundred and fifty horse race betting markets on Betfair and Betdaq, I am moving towards the notion that in the case of these two betting exchanges, bid-ask spreads are pecuniary externalities; that is, a class of externality that does not constitute market failure. I have observed a number of situations in which liquidity, measured in terms of depth (the size of the bet on offer), tightness (the spread between the back and lay sides) and immediacy (the availability of the bets on offer), is signifciantly worse on Betdaq's markets compared with Betfair. However, in terms of total bets matched, Betdaq actually outguns its rival. There may be any number of reasons for this, but what I am beginning to observe, is that the majority of this money is being matched on the favourite in each race, at the expense of the rest of the field (on which the spread between the back and lay sides is large). This would suggest that the largest traders on Betdaq, are big bookmaking companies, who are using the exchange to offload liabilities that they have on heavily backed favourites. The Betfair exchange, whilst populated by some bookmakers, is also home to thousands of traders, who are actively involved in spreading liquidity across all the runners in each race. However, as on Betdaq, the favourite in each race still accounts for a significant proportion of the total bets matched. Having compared two hundred and fifty horse race betting markets on Betfair and Betdaq, I am moving towards the notion that in the case of these two betting exchanges, bid-ask spreads are pecuniary externalities; that is, a class of externality that does not constitute market failure.

I have observed a number of situations in which liquidity, measured in terms of depth (the size of the bet on offer), tightness (the spread between the back and lay sides) and immediacy (the availability of the bets on offer), is signifciantly worse on Betdaq’s markets compared with Betfair. However, in terms of total bets matched, Betdaq actually outguns its rival.

There may be any number of reasons for this, but what I am beginning to observe, is that the majority of this money is being matched on the favourite in each race, at the expense of the rest of the field (on which the spread between the back and lay sides is large). This would suggest that the largest traders on Betdaq, are big bookmaking companies, who are using the exchange to offload liabilities that they have on heavily backed favourites.

The Betfair exchange, whilst populated by some bookmakers, is also home to thousands of traders, who are actively involved in spreading liquidity across all the runners in each race. However, as on Betdaq, the favourite in each race still accounts for a significant proportion of the total bets matched.

]]> By: Byrne Hobart http://www.midasoracle.org/2007/10/23/network-effects-network-externalities/#comment-16376 Byrne Hobart Wed, 24 Oct 2007 15:46:44 +0000 http://www.midasoracle.org/2007/10/23/network-effects-network-externalities/#comment-16376 Has anyone formalized network effects to account for how the early adopters probably have higher marginal utility? I would expect that the first practical phones (used, I think, to rapidly re-route trains) added a lot more wealth per phone than, say, a recently-purchased cell phone. Has anyone formalized network effects to account for how the early adopters probably have higher marginal utility? I would expect that the first practical phones (used, I think, to rapidly re-route trains) added a lot more wealth per phone than, say, a recently-purchased cell phone.

]]>