Natural monopolies and vertical integration December 3, 2012 at 11:17 am

The SWP has a post here on the on-going swap data repository mess (SDR) in the US. I admit to not having fully thought through the pros and cons of vertical integration in this space, although I do have concerns about the monopoly issues that come with vertical integration, so I won’t comment on the first part of his post. This, though, is clearly right:

there are substantial scale and scope economies in creating and operating an SDR, and that an SDR is a natural, multi-product monopoly. It is highly likely that the efficient form of organization would be an industry utility/cooperative, likely a non-profit.

Unfortunately that is not what we are going to get. Therefore, as SWP says, “although mandatory reporting to a [single] data hub/SDR is a good idea, as a result of political economy considerations it will be implemented in a highly inefficient way.”

2 Responses to “Natural monopolies and vertical integration”

  1. Thanks, David.

    One quick comment re “monopoly issues that come with vertical integration.” This is the gravamen of the paper I gave at Berkeley. It’s a question of the direction of causation. Implicit, I think, in your concerns is a belief that vertical integration causes monopoly, or at least is a way of foreclosing competition. In my paper, I argue that the causation goes the other way. Due to the strong scale economies in execution in order driven markets, clearing, and settlement, each of these is plausibly a natural monopoly, or at the very least a highly concentrated natural oligopoly. Vertical integration is a well-known way to mitigate the contracting problems (opportunism, holdup, inefficient coordination) that arise when separate, vertically related monopolies deal with one another

    The paper looks at the foreclosure issue theoretically and empirically, and concludes that vertical integration is an efficiency enhancing response to monopolies that arise due to the nature of execution and clearing.

    The inherent problem with trading markets is that fundamental economic considerations imply that they will almost never be even modestly competitive, except in spurts when an entrant competes to supplant the incumbent monopolist (e.g,. Eurex supplanting LIFFE in Bunds in ’98). Vertical integration is a response to that, not a cause of it.

  2. Thanks Craig. I hadn’t thought of that before, I have to confess. I’ll try to track down your paper: the dynamic you describe is certainly interesting. I don’t quite see your point that “Vertical integration is a well-known way to mitigate the contracting problems that arise when separate, vertically related monopolies deal with one another”. As I understand the Coase view, vertical integration is often an efficient response to contract friction between upstream and downstream, so I don’t understand how it relates dealings between vertical monopolies…