Ian Cook

[NOTE: Our anti-spam software does not like something in Ian's comment, so he graciously gave his consent to post this as a guest post -- ed.]

Unfortunately, I don’t know much about the institutions specifically within the electric power industry, so I imagine this will be of limited help. But, that’s never really stopped me from rambling on.

Ostrom was one of the first things that came to mind after reading through your posts. Though, the book I had in mind was Rules, Games, and Common Pool Resources. Specifically, some of the discussion in the book about the levels at which rules are defined; in the context of altering institutions, she defines “constitutional level” rules that define the boundaries around who can make choices at a collective level. It would seem to me that any serious change would not only come from that level, but have to affect the actions on that level. If the ability to make change at that level is blocked then I would think there would only be a limited space in which the operations of the institution might be able to fluctuate. In my simplistic understanding of the potential principal-agent relationship, the institution would thus simply generally reflect the governing body or group that has control over the constitutional-level rules. The more resistant to change that group is, the less likely real change will be. To discuss the potential for change, Chapter 12 of RG&CPR talks about the potential movement from CPR usage being a PD-style game to an assurance problem-style game. Again, the question of monitoring and information sharing aids this, as you’ve mentioned before.

While it may not be at all applicable, I guess I saw some of this in terms similar to that discussed by Krehbiel in Information and Legislative Organization insofar as one of the problems in institutions (such as Congress) is a tension between the ownership of information and the ownership of the ability to enact change through rule-making (of course, agenda-setting issues also play a massive role). That the group with specific information has a suggestion, and is supposed to be some sort of representation of the the whole institution, is no assurance that their suggestions, whatever their instrinsic value, will be adopted. Perhaps in the case of institutions even outside of congressional control, there is a similar problem of walls existing between those with policy-specific knowledge and those with institutional-change prerogatives. (Perhaps this is blindingly obvious. Just don’t know.) The focus of the bureacrat would be the cost of the operation, rather than on the policy output issues of interest to the policy-specialist.

Though it may be a poor comparison, the differences in approach towards cellular service in Europe and the US came to mind when reading these posts. Europe chose to have a standardized format for cellular service; they opted for uniformity and interoperability. The result was limited ability to react to massive growth in bandwidth usage, higher prices (since the competition was for the phone, not the service), and slower development of technology that pushes the limits (which, I think, is why Europe was always so eager to hear about the “next release” of phones, such as 3G and whatnot). By contrast, the US didn’t standardize. We have lower rates for service, a faster adoption rate for new technology (CDMA, TDMA, etc. and so on), but far less interoperability. We can’t yet buy a soda from a vending machine or pay for metro rides using our phones — a rising practice around the world. to make the connection, the various competiting US service companies understood far better the potential rewards of pushing the service technology, with little governance at the top to tell innovators not to follow certain roads. The European system created a structure that relied on bureaucrats to choose, with their limited knowledge, and did not allow those with specific information to do more than “provide guidance”. I think some of that may be tied to the duration off those in the controlling group. The beneficial increasing returns from having the institution are “misread”, perhaps, and longer-view benefits aren’t counted in. Roll-out in Europe was certainly faster because service was standardized. But the longer-view benefit of faster progress could have been seen, perhaps, as more valuable had not the bureaucrats been more concerned with what happens during their tenure. (Akin to much of what Douglass North has in Institutions, Institutional Change, And Economic Performance.)

If some of these issues do involve US federal institutions, maybe there’s something of value in Paul Light’s work on the “thickening of government”. The pool of people at the top of the institution is getting a lot heavier. That is, the people with constitutional-level control grows and thus the number of people depending on the status quo grows; more importantly, the ratio of those with the ability to change the rules to those who have specific knowledge starts getting much smaller. I can’t imagine that could be good for future change.

Lastly, I suppose another lens might be that of Hart and the question of complementary versus independent assets. Though, in this case, it might be a function of counter-argument. The internet is “controlled” by a loose affiliation of groups; the assets strike me as highly complementary with my cursory knowledge of their various jurisdictions. We’d tend to believe, if the Hart position was applicable in this case, that they would realize efficiencies from organizing into a single “firm” — yet they haven’t, and actively work to prevent such a move.

Lynne Kiesling

In my previous post about organic institutional change, I pondered Ostrom’s eight institutional design principles. She applies her analyses to common pool resources (CPRs). Are they also applicable to changing electricity regulation institutions?

Obviously I suspect that the answer is “yes” or I wouldn’t be putting us all through this. One of the defining features of electric power networks, as with CPRs, is interdependence of individual actions. Going point by point through the first three of the eight design principles, I’m going to elaborate on how I think we can apply them in electric power.

1. Clearly defined boundaries. Ostrom considers specifying boundaries and use rights as “a first step in organizing for collective action.” (p. 91) The ability to define boundaries in an electric power network depends on some of the physical features of electric power networks. When we take into account two aspects of transmitting power on a network – Kirchoff’s Law and the interdependent nature of reliability, voltage, and power quality across agents – we get a clearer idea of the parameters of boundary definition here and its limitations. Electrons following the path of least resistance (Kirchoff’s Law) mean that a contract path and the physical path of the power are unlikely to match; the interconnections of agents on the network mean that all interconnected agents will experience some common features of the network. Physically excluding agents is difficult, which means that the ability to define “rules limiting appropriation and/or mandating provision” (p. 92) will be really important in electric power networks.

Use rights in electric power networks are more complex than for Ostrom’s irrigators. In that case there are n agents with rights to use a resource of uncertain and changing size, and those agents use by appropriating or extracting. In electric power, there are k sellers of power, m buyers of power, and n owners of the CPR that the (k+m) buyers and sellers are using. n can vary from 1 upward depending on the region and the distance covered in the transactions. The size of the CPR is more or less known, depending on the timeframe – future use and investment are uncertain.

The past decade has seen the development of transmission rights (FTRs) and efforts to work through the definition of use rights on the electric power network. But FTRs are but one part of defining use rights on the network. There is another use right that is implicit and not clearly defined – the use rights of the m buyers. In the FTR designs and transactions, the participants are the (k+n) sellers of power + owners of the network. But treating the m buyers of power as “load” implicitly gives them unrestricted use rights. Is it any wonder, then, that we see use patterns on the grid that look like overuse of a CPR?

Perhaps underinvestment in the network is not as much of a problem. Perhaps the primary problem is that the m buyers have unrestricted use rights, and our regulatory institutions do not allow the (k+m+n) full set of agents to devise appropriate rules to govern the use of all of the agents, not just the supply-side ones.

2. Congruence between rules and local conditions. In the US the electric power network comprises “broadly similar, but distinctly different, systems.” (p. 92) Contrary to common belief, local physical conditions do vary on the grid (which has implications for the degree of interdependence of agents). Also, local conditions vary across the country according to geography, economic activity, and political environment. Crafting well-tailored rules that reduce transaction costs across an entire network but still incorporates and respects local conditions is a daunting challenge in any CPR environment, and has been one of the struggles in the Standard Market Design debates in electric power in the past three years. I think a lot of the hard thinking and heavy lifting has to happen on this principle.

3. Agents can participate in modifying the rules. Allowing a process for agent modification is a powerful tool for exploiting local knowledge and forward-looking entrepreneurship. Being able to change the rules as agents observe whether ex ante commitments to abide by them are borne out ex post contributes to crafting rules that are more robust. This tool increases the use of local knowledge, decreases monitoring and enforcement costs, and increases the probability of compliance, which increases certainty and stability for all agents. It does come at a cost though; there will certainly be transaction costs to modifying rules. But the net effects of these costs and benefits have to be weighed against the net effect of the substantial and varied costs that we bear under the current set of rigid, inflexible institutions.