Lynne Kiesling
Here's an interesting Technology Review interview with Digg founder Kevin Rose, talking about about the new ideas and creativity that flourish when you just define the platform and let the participants create the content. A good complexity and emergent order read.
Michael Giberson
Thoughtful, useful, inadequate and wrong-headed. That’s my reaction upon completing Richard Bookstaber's book, A Demon of Our Own Design: Markets, Hedge Funds, and the Perils of Financial Innovation.
As I mentioned last week (in The Perils of Financial Innovation, or Not), Bookstaber was among the MIT-trained economists lured from academia to Wall Street in the 1980s. In the book he gives an insider’s account of the development of financial innovations in the 80s and 90s, linking the innovations to the 1987 market crash, the failure of LTCM, and sundry less notable blow ups. For my taste, there was too much tracking of who did what where, and too little explanation of how various financial mechanisms were supposed to work, and why they sometimes went bad. About midway through the book, Bookstaber shifts emphasis from careers and organizations to his diagnosis of problems that underlie the system and a few ideas on how things could become better.
Overall, Bookstaber failed to convince me that “less complexity and looser coupling of markets” is a reform banner that I should be willing to march under. Bookstaber himself explains clearly the futility of hedge fund regulation, so it is less than clear what sort of public policy approach is warranted. (While last week I expressed concern that he might advocate reforms that simply shift risks back into the real economy, he avoids that temptation.) Individual investors, and particularly the firms that manage investments, may find some wisdom in Bookstaber’s hard won understanding of the limits of computational techniques – but this book is hardly the only place for finding these insights.
So the book is useful as a report on the financial innovations of the 80s and 90s, and inadequate and somewhat wrong-headed in his suggestions of a reform agenda. But my final reaction is in the word I led with: thoughtful. While I’m not convinced by Bookstaber’s talk of normal accidents and tight coupling, the book does offer a deeper understanding of working of financial markets and in particular of the contributions of hedge funds to “greasing the wheels of capital.” In a world where financial innovation continues to reshape markets and therefore the economy, deeper understanding of financial markets will prove invaluable to economists, business managers, and particularly policy makers.
Lynne Kiesling
Mark Thoma draws our attention to a recent Martin Wolf column in the FT, discussing complexity and evolution as a methodological framework for economic modeling and analysis.
[M]ost [economists] must also know that the economy is not characterised by perfect foresight and equilibrium, but by trial and error and evolution. That was the intuition of the Austrian economists, Joseph Schumpeter and Friedrich Hayek. But this vision has had next to no influence in the discipline itself.
This observation is a lead-in to discussing Eric Beinhocker's The Origin of Wealth: Evolution, Complexity, and the Radical Remaking of Economics. Mark's post has useful excerpts from Fox's column.
I have not yet read Beinhocker's book, but I think his point about economic methodology is important. Models premised on perfect foresight and completely knowable information sets, and focused on outcomes instead of processes, are going to do a bad job of helping us to understand economic behavior and phenomena that are complicated, like growth, innovation, and entrepreneurship. I also contend that they do a bad job of helping us to design regulatory institutions that can adapt to changing and unknown conditions. That's why I pay attention to complexity science, because I think it can help us build more useful models of institutional change.
Lynne Kiesling
For those of us interested in the economics of things like "mass customization", the department store is a fascinating phenomenon. A retail business model originating in the late 19th century, the department store for decades epitomized elegance, convenience, ubiquity of options. Then in the 1990s the department store fell on hard times as nimble, smaller retailers struck better production and/or procurement contracts, had more direct contact with the preferences of consumers, or were able to offer niche products to enable consumers to craft their own, individual, modern images.
Now Grant McCracken tells us that the department store is back. He cites J.C. Penney's success in the mid-market range, which is correct. Here's his summary of why the specialty store has been succeeding for the past 15 years:
The usual explanations for this decline are vanishing sales staff, badly organized stores, and fashion insensitivity. I think there was another explanation that didn't get enough attention. The specialty retailer was a better meaning manager.We can chart the decline of the department store against the rise of the national brand. As branding got better, and marketers became more skilled, the department store became more punishing. It was so uninviting, so unorganized, and so aesthetically unforgiving, even the best brands began to wilt.
A response was inevitable. Ralph Lauren said, "leave this to us," and build little boutiques into the department store. These boutiques out-earned the rest of the floor because they continued to build the brand. Mr. Lauren's store was a bastion of privilege in what was otherwise biggish, boxish and artless. I heard, but never confirmed, that Mr. Lauren had a full time staff member to search out those "rowing team" photos that gave the store it's preppy feel. The boutique could do meaning management that the department store hadn't known since in it's golden palace hey day.
Yes, department stores are doing better financially now, as they streamline their supply chains through merger (Macy's) and correct some of their recent missteps (Saks). Nordstrom and Bloomingdale's have been more steady and have not suffered some of the setbacks that drove stores like Bonwit's and B. Altman out of business.
Grant thinks the department store may even be able to do the specialty stores one better:
Ah, this is interesting. The specialty store could go deep. It could cultivate the brand carefully and well. But in a hyperactive marketplace, where consumer taste change often and shifts suddenly, the real challenge is remains current. And it is easy to do this with many brands supplied by other players than one perfectly managed brand of one's own. Retail, a river runs through it! This is it's adaptive advantage. Potentially, the department store can be a complex adaptive system.
But I'm not convinced that the department store is back. I still think the specialty store does a better job of managing meaning and selling image. I am not convinced that the large department store that is managing many brands and a national image can be more nimble than a specialty store, and nimbleness is what a department store will require to become a successful complex adaptive system. I don't think traditional retail management will be willing to tolerate distributed control, which is what you need to have a successful complex adaptive system.
Lynne Kiesling
Over at Organizations & Markets, Peter Klein and Nicolai Foss have been talking about tacit knowledge, distinguishing it from explicit knowledge, and asking what the big deal is. Their posts and the comments on them are a good read.
Why should tacit knowledge be any more or less important than explicit knowledge? My sense is that some of the researchers Peter and Nicolai mention emphasize tacit knowledge not because it is superior to explicit knowledge, but because it is so overlooked in social science research, and particularly in economics. Tacit knowledge defies formal modeling, and is a somewhat difficult concept to grasp: to be all Donald Rumsfeld-y about it, it's stuff that you don't know that you know. It's also stuff that you know how to do that you never consciously learned. But that's one reason why experiments are a valuable social science and policy tool: people can't articulate their tacit knowledge, researchers can't know in advance what tacit knowledge to look for or how it will affect outcomes, but when you observe behavior and interactions in controlled environments, tacit knowledge has an opportunity to act upon decision-making.
Lynne Kiesling
Today is a dynamism day here at KP, whether it's marketing or technology or business models. Business Week has an interview with Frank Moss, the new head of MIT's MediaLab. He points out the importance of several things that I think are crucial for dynamic innovation: entrepreneurship, new models of collaboration and invention, and focusing on disruptive change instead of incremental tweaks:
How do you view the nexus between technology and entrepreneurialism?
It is hugely important. In fact, entrepreneurs are really the primary vehicle for innovation in our society. They've played an incredible role. Thirty years ago, the primary source for innovation was large corporate labs. That is where all of the money went. Then, 20 to 25 years ago, the source of ideas and creativity shifted to venture funds and startups.Over the past 20 years, we've seen the economy and society change due to innovation from small independent efforts outside of corporate labs. Technology has enabled startups to have a big influence, and consequently they have had a tremendous effect in the technology scene today.
What role will startups play in the future?
I see tremendous economic growth from startups from 10 years ago. Entrepreneurs will go from the 1,000 startup ventures funded in the last 10 to 20 years to ideas coming from people working together in network-based environments, using computers to dream up innovations in a way they never did before. It could be people in developing countries with low-cost computers.The Media Lab has given a start to many entrepreneurs. What would be your advice to would-be entrepreneurs in today's environment?
Resist the current temptation to make incremental changes to attract funding. It might get you off the ground, but I don't think it will get you very far. Today, the funding climate has changed. The successful (entrepreneurs) will look for fundamental disruptive change. I encourage them to take risks, rather than just polish the faucets. There will always be an appetite for game-changing technology.
Note in particular his focus on distributed, decentralized, networked innovation. This way of thinking about innovation taps in to the idea that creativity resides in many distributed individuals, and that a distributed network may do a better job of generating beneficial disruptive change than a centralized, coordinated, top-down lab approach.
I think it makes for an interesting challenge for an organization like MediaLab; at some level, it will be harder to direct and target resources on projects that are likely to be transformational, because so much more innovation will be distributed and not concentrated in big labs. At another level, though, MediaLab can seed particular projects or individuals who are working in a distributed network, and that bottom-up approach might be more likely to lead to successful disruptive technologies than the coordinated large-lab approach.
Thanks to Glenn Reynolds for the link.
UPDATE: Paul at Truck and Barter has a link-filled post on the same question, inspired by an article in the Economist on a recent AEA session on entrepreneurship. Note his link to Chris Coyne's post at Austrian Economists, where he laments the absence of attention to Kirzner in the Economist article, a lament that I second.
Lynne Kiesling
Am I really the first person in my online reading network to notice that Malcolm Gladwell's got a blog?
Yee hah! There are a few people whose trenchant observations make me more productive, and he's high on the list.
Lynne Kiesling
Thanks to Ian at Truck and Barter for his pointer to Complexity Blog, from two Michigan graduate students. They are certainly hitting on some topics that overlap with our interests here at KP.
Lynne Kiesling
Steve Horwitz's musings on complexity and snowflakes is a great read that raises an important question: why are so many of us willing to accept and appreciate emergent order processes when they occur in non-human nature? Why do some think differently about snowflakes in the air and eggnog in the store?
People who are so willing to accept the existence and beauty (and benevolence!) of undesigned order in the natural world should be more willing to open themselves to the possibility that there are processes of undesigned order at work in the social world too.
Both snowflakes and eggnog are seasonal occurrences, and both are products of complex processes about which we can know almost nothing, yet still we can enjoy and appreciate the product of the process. But we also struggle with the desire to impose control over those social processes.
Emergent order in human processes is as beautiful as physical emergent order. This observation dovetails with Pete Boettke's Mystery of the Mundane post, particularly Pete's recommendations to students:
Allow yourself to be curious about the world, and to find the mystery in the mundane. Buying a cucumber at the local grocery shop, the shirt on your back, and the shoes on your feet are all fascinating subjects to explore using the economic way of thinking. Finding the hidden pattern in the buzz of daily life is what we will do in the first 6 weeks of class. If you allow yourself to be amazed by the world around you, then you will do well in studying economics.
Thanks to Alex Tabarrok at Marginal Revolution for the link.
Lynne Kiesling
I spent last weekend attending a Liberty Fund conference on themes of liberty and responsibility in the writings of Charles Darwin. We read selections from The Origin of Species, The Descent of Man, The Expression of Emotion in Man and Animals, and Darwin's autobiography. It was a fascinating opportunity to learn about Darwin's theories, models, methodology, and data on his own terms.
From an economics perspective, the value of the discussion was the opportunity to think more specifically about how we use metaphors of evolution and natural selection in our analyses of institutional, social, and technological change. The simple metaphor is pretty straightforward. Through a distributed, decentralized process (natural selection in Darwin's model), agents that behave in ways that are favored in their environment exhibit a higher degree of fitness, and thus have a higher probability of persisting. The traits or behaviors that lead to that fitness in that environment persist while others recede, leading to incremental changes in the characteristics of the agents as they adapt to the environment. When the environment changes, which happens almost constantly and imperceptibly, sufficiently dramatic changes will change the relative fitness value of different traits or behaviors, leading to further adaptation. This ongoing process creates the variation and complexity that we observe in the natural world. It also allows for rogue beneficial mutations to flourish, and for malevolent mutations to be crushed. In this way the process has some global stability properties, notwithstanding the constant change.
How good a metaphor is this for institutional, social and technological change? I like it and think it's a pretty good one, although clearly it's not a true 1:1 mapping from biological variation and changes in agents (individuals or "species") into the institutional sphere. One of the mechanisms of natural selection in biology is sexual selection, and I think it's a stretch to apply sexual selection processes to institutional change. But if you model technological change as either a process of adaptation or beneficial mutation (or both), you can get pretty far with the metaphor. Evolutionary economics and evolutionary game theory have done just that.
I think the metaphor for institutional change is less exact but still useful. I had been trying to model institutions and the process of institutional change in the way just described for technological change, but it has been ringing a little hollow to me. I like thinking about institutions as agents that change in adapting to their environment, because that gives me a theoretical foundation for using what I call "the evolutionary criterion" to evaluate policy changes: is the institution sufficiently flexible to adapt to unknown and changing conditions? This way of modeling the problem requires thinking about the environment as being the economic and technological environment, with policies as agents in the environment. I can also think of particular institutions as having different levels of fitness. It gets me what I want, but sounds a little contrived when I try to formalize it like this.
So what if I think instead about institutional change as changes in the environment, to which agents (individuals, firms, consumers, producers, regulators) adapt to varying degrees? Still interesting, but a different model, a different game, leading to different hypotheses and policy recommendations. And I think that's been done, and done well, by folks like Paul Rubin in his book Darwinian Politics. This way of structuring the problem also works well for answering public choice questions, rent seeking, etc.
But my primary question of interest is how we design institutions to maximize their fitness in a dynamic economic and technological environment. I think that means thinking of the institutions as agents interacting with their environment (and presumably with each other, to complicate the model).
Note that I used the D-word: design. Natural selection is a bottom-up process, in which variations in agents emerge in response to the requirements implied by the environment and its changes. But institutions are not natural. At the same time, though, can we think of and design institutions that are not entirely artificial? Can we design institutions that are more approximately natural than artificial? And can we do it in a way that has these good fitness/efficiency characteristics, as well as some global stability properties? That requires flexibility and adaptability, yet also at the same time transparency. That's the challenge.
Lynne Kiesling
Cass Sunstein is currently guest-blogging at Lessig's place, and has a thought-provoking post on information aggregation, Wikipedia, prices, and Hayek. The post itself and the comments have a lot of good content, and I recommend reading them. It also follows on a prior post on information aggregation more generally, as Sunstein works on a follow-up to his Republic.com book.
One of the interesting things to me is that in the comments on the earlier post, Jimbo Wales (Wikipedia founder) says that Hayek's ideas informed how he designed Wikipedia, which I had not heretofore realized.
I have several, perhaps random, thoughts on all of the ideas raised in these two posts and the comments therein.
One issue raised is the relevance of the comparison between prices as information aggregators and wikis as information aggregators. It's a compelling comparison, but I agree with one of the commenters who argued that the comparison to prices is better as a metaphor than as an analytical tool. One reason it's a better metaphor than analytical tool for the wiki process is that market processes and the prices that emanate from the interaction of market participants are more complex and not so specifically goal-oriented to aggregation. One commenter made this point well: prices are a manifestation of a decentralized mechanism for facilitating the mutual benefit of buyers and sellers; they do not come into being because anyone has the specific goal of aggregating information.
Prices come into being because they are a parsimonious and useful tool for coordinating the wishes of diverse, heterogeneous individuals into mutually beneficial actions. The information aggregation is an important and useful part of that process, because it's the aggregation of private information on preferences and costs that lets you figure out if you want to buy or sell at that price. Through that process you discover how strong your preferences are, or how high or low your costs are, so you learn the information about the preferences and costs of others in the process of figuring out if it's worth it to you to buy or sell at that price. But the aggregation is not the goal of the process; the goal of the process is individual happiness, for both buyers and sellers.
And the beauty of prices is that they coordinate not just the diverse, private information and incentives of individuals, but in so doing they enable those individuals to create and receive value. Gains from trade. Surplus. Happiness. Mutual benefit. One of the commenters claimed that wikis are different from prices because in markets it's about individual gain but in wikis it's about social gain (my paraphrase). That is a naive misunderstanding of market processes. Market processes are robust and valuable precisely because they provide an institutional framework in which individual incentives are, through no conscious intention, channeled into creating social gain. This is one of the fundamental and oft-overlooked insights about markets: in order to have exchange, you have to cooperate (by adhering to rules/laws) and you have to have something to offer that others want, so you have to pay attention to the desires and preferences of others. As a seller you cannot gain if you are offering stuff that no one wants, or at prices they aren't willing to pay, so you have to pay attention to someone's preferences other than your own.
As a small point because this post is long enough, I would also second the remarks of the commenter on the second post who said
You claim that Hayek was "too optimistic" because he didn't understand that the "price system doesn't always work". This is a misreading of Hayek. Can you provide a citation in which he claims that the price system "always works"? I suspect not. Nothing "always works," and Hayek was smart enough to realize that.Hayek's claim is that the price system works better on average than any other system.
Yes. One of the valuable dimensions of Hayek's contribution to understanding human action is the recognition that human institutions, including markets, are not ever and cannot ever be perfect, because of the cognitive limitations of humans. However, he made the more defensible claim that among the class of human institutions, markets perform better on average than other institutions for the allocation of goods and services and the bringing into being of new goods and services. And of course, rules/institutions matter in determining how well market processes perform those valuable roles, which is why he devoted so much of his work to analyses of emergent order legal systems and their ability to provide necessary monitoring and enforcement to sustain the decentralized network systems we call civil society.
Thanks to Orin Kerr at Volokh for the tip.
Lynne Kiesling
Russ Roberts has an absolutely wonderful essay at Caf Hayek. Russ uses a recent FedEx transaction, and how sure he was that his package would arrive unmolested and on time, as a starting point:
My lack of worry at FedEx is different. It can't be trust. There are too many people touching my package who I will never see. How can I trust them? I know nothing about them. The woman behind the counter seemed like a decent enough soul. I trusted her somewhat. But I don't even think that's the right word my feelings about her. But it's certainly the wrong word to describe her co-workers who brought my package safely to St. Louis. I can't say I trusted them. I knew nothing about them.Faith? Seems too open-ended or something. Faith comes from having used FedEx before and knowing that they always get the job done. There's a little of that. But I wasn't even worried the first time I used FedEx.
Confidence seems like the right word. Confidence born from an understanding of how the division of labor works in a modern economy. What Hayek called the extended order of human cooperation.
The essay is too rich to excerpt well, so please read the whole thing. The important point is that in complex human interactions, we develop different orders to help us structure the complex world given our physical and cognitive characteristics. A relationship and method of transacting that works for getting the kid next door to mow your lawn is unlikely to work for getting a parcel from Washington to St. Louis. And more importantly (I think), that ability to create different types of extended order doesn't only enable us to manage complexity and survive notwithstanding the complexity: it enables us to use complexity to our advantage.
This ability to use complexity to our advantage is one of the hallmark characteristics of two very important market institutions: prices, and long-distance, impersonal exchange. Diffuse, personal knowledge is a dimension of complexity, and we take advantage of it through the price system because we need only know the price facing us individually in order to make the decision that's right for us; we need not know the preferences or costs that went in to the process of the price getting to that point. And with impersonal exchange we get to leverage the heterogeneity of both our preferences and our costs, as well as the heterogeneity of opportunism and entrepreneurship, to exchange goods and services in ways that we could not if we limited ourselves to exchange through a more intimate, personal order.
One of the most fruitful areas for entrepreneurship (both commercial and institutional), and for research, is in investigating how we can use complexity to our advantage and create new ways of ordering our lives that use those advantages, instead of just thinking that we have to cope with complexity and to control and manage it. Instead, embrace the complexity.
[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 Ostroms 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 wouldnt 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, Im 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 Kirchoffs 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 (Kirchoffs 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 Ostroms 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.
Lynne Kiesling
In my previous "thinking out loud" post on institutional change, I ended with this question:
Institutional change is in many ways itself a constructivist exercise. Is there a way to make the process of institutional change more organic, and thus more likely to lead to "valuable, meaningful, forward-looking, robust, evolutionarily adaptive institutional change"?
For help in thinking through this question I turned to one of the most important works about institutional change, Elinor Ostrom�s Governing the Commons. Ostrom focuses on how communities develop self-governance institutions for the mutually beneficial use of common pool resources (CPRs), particularly water and land. Ostrom's work has shaped my thinking on these issues, as is evident in the property rights paper that I co-authored with David Haddock in 2002.
Ideas about institutions governing CPR use are highly relevant to electricity policy, but have not yet been applied. At a very fundamental level, CPRs and physical networks share crucial economic features, most notably the interdependence of both production and consumption decisions of individuals. In her CPR context Ostrom calls individuals "appropriators" because she is focusing on, for example, the appropriation of water from a shared river; in a network context let's call those individuals agents. In both cases, appropriators and agents make decisions based on the institutional environment in which they operate, which influences their incentives and opportunity costs. Those decisions can and will affect the consumption and production of other agents, and in a dynamic sense will also affect their incentives for how much to consume and produce in the future. This is the sense in which the analysis of so-called public goods and externalities using the prisoner's dilemma is so powerful.
Traditional arguments about how to get out of the prisoner's dilemma in CPRs involve privatizing the commons. In electric power, this strong property rights approach is not feasible because of the network nature of the transactions and because Kirchoff's Laws mean that the physical flow of electrons and the contract path corresponding to assigned property rights are not likely to match up. So Ostrom�s exploration of situations in which people choose to keep a CPR as a CPR but use governance institutions to get out of the prisoner's dilemma offers some insights about what some of the features are of robust institutions.
In Chapter 3 of Governing the Commons, Ostrom distills some design principles from case study analysis of several long-enduring CPR institutions (Table 3.1, p. 90). She does not claim that these design principles are either necessary or sufficient, but that experience of robust CPR institutions suggests that these design principles increase the probability of success.
1. Clearly defined boundaries: rights to withdraw resources from the CPR must be well defined. Note that this principle relies on a very crucial distinction in both CPRs and networks: the distinction between property rights and use rights. Either by choice or physical necessity a group of agents can treat a resource as a CPR while still defining very specific, clear, transparent use rights. This principle is very important in a network context.
2. Congruence between appropriation and provision rules and local conditions. Local conditions matter, a lot. We have millennia of human history, with the most poignant and painful being the Soviet experience, showing how the top-down imposition of external institutions that do not respect or incorporate local knowledge and local conditions will make people worse off and are ultimately doomed to costly failure. This design principle gets at the core of my question by indicating that robust institutions are organic in nature; they derive from the decentralized interactions of those with strong interests and good local knowledge. Local knowledge and local conditions differ, even on an electric power network. Furthermore, some of that local knowledge is going to be tacit and difficult or impossible to communicate, reinforcing the important points of Hayek's "The Use of Knowledge in Society".
3. Collective-choice arrangements. Agents affected by the institutions should have opportunities to modify the institutions. Otherwise the institutions stagnate as networks evolve, and the institutions become obsolete at great cost (both financial cost and opportunity cost) to the network agents.
4. Monitoring. Monitoring is important but costly, and should be done in a way that the monitors are accountable to the network agents. In many of the CPR cases Ostrom analyzes appropriators themselves do the monitoring, although it's important to have a third party responsible for enforcement and punishment. In a network context, where everyone's actions are to some degree interdependent, self-monitoring arrangements should be relatively cheap, but getting the incentives right for the monitor is a challenge.
5. Graduated sanctions: punishment escalates with the severity of the violation.
6. Conflict-resolution mechanisms: these can be formal or informal, from courts to local self-administered arbitration. In different contexts different conflict resolution mechanisms will have the credibility to make their decisions stick, so this design principle also builds on the local conditions/knowledge one discussed above.
7. Minimal recognition of rights to organize: the rights of local agents to develop and arrange their own institutions are not challenged by central government authorities. This one is a toughie in a network context like electric power that occurs in multiple political jurisdictions, and with this history of 85 years of symbiotic codependency of the regulator and regulated. But it's an important one, because this is a pressure valve that allows for the evolution of institutions and the adoption of designs that are better suited to local conditions and/or changing and unknown conditions. Government-based institutions do not typically face competition from different institutional arrangements, but that competition is crucial (and is one of the important foundations of the "laboratory of the states" defense of federalism).
8. For more complex or heterogeneous populations or resources: Nested enterprises. I interpret "nested enterprises" as analogous to the concept of federalism, with local knowledge and local institutions prevailing where appropriate but nested within an institutional structure to accommodate larger and broader interests.
These design principles can inform how we design institutions in electric power that have more organic features, and are more likely to be flexible enough to evolve beneficially in the face of unknown and changing conditions.
Please share your thoughts with me about how these design principles are relevant in electric power, and any ideas you have about implementing them.
