We consider how the introduction of centralized netting in financial networks affects total netted exposures between counterparties. In some cases there is a trade-off: centralized netting increases the expectation of net exposures, but reduces the variance. We show that the set of networks for which expected net exposures decreases is a strict subset of those for which the variance decreases, so the trade-off can only be in one direction. For some network structures, introducing centralized netting is never beneficial to dealers unless sufficient weight is placed on reductions in variance. This may explain why, in the absence of regulation, traders in a derivatives network do not develop central clearing. Our results can be used to estimate margin requirements and counterparty risk in financial networks.

The value of Bitcoin depends upon self-fulfilling beliefs that are hard to pin down. We demonstrate this for the case where Bitcoin is the only form of money in the economy and then generalize the message to the case of multiple Bitcoin clones and/or a competing sovereign currency. Some aspects of the indeterminacy we describe would no longer hold if Bitcoin were an interesting-bearing object.

Experiments on second-price sealed-bid private value auctions have established that subjects typically bid more than their value, despite the fact that value bidding is a dominant strategy in such auctions. Moreover, the laboratory evidence shows that subjects do not learn to bid their values as they gain more experience. In the present paper, we re-examine the second-price auction data from Kagel and Levin’s (Econ J 103:868–879, 1993) classic paper. We find that auction efficiency increases over time, even though the frequency of overbidding is unchanged. We argue that the rise in efficiency is due to a decline in the variability of overbidding. This is consistent with subjects’ learning to bid more like each other.

Systemic risk among the network of international banking groups arises when financial stress threatens to crisscross many national boundaries and expose imperfect international coordination. To assess this risk, we use Rosvall and Bergstrom’s (PNAS, 2008, 1118-1123) information theoretic map equation to partition banking groups from 21 countries into modules. We consider a quarter of a century of data on the cross-border interbank market. We show that in the late 1980s four important financial centres formed one large super cluster that was highly contagious in terms of transmission of stress within its ranks, but less contagious on a global scale. But the expansion leading to the 2008 crisis left more transmitting hubs sharing the same total influence as a few large modules had previously. We show that this greater entanglement meant the network was more broadly contagious, and not that risk was more shared. Thus, our analysis contributes to our understanding as to why defaults in US sub-prime mortgages spread quickly through the network.