When a tennis player decides if and when to use their rights to ‘video review’ of points they are trying to solve cognitive and tactical problems. When a cricket captain decides to review an umpire’s decision there’s an additional problem. Challenges have been rationed by design to prevent frivolous challenges. But the individual player with the best cognition – the batsman or bowler – has a strong interest in influencing the result. Indeed, so strong that it could even warp their cognition, let alone their judgement. So how should the captain make the decision?  The captain has a governance problem.

To generalise, a governance problem exists where the ‘brain’ or decision making of some entity – a biological or social organism – must access information from within that organism to make a decision and there is no unproblematic way of generating and/or handling the information. These kinds of problems are ubiquitous. There are lots of governance problems to be dealt with in jellyfish and vastly more in mammals. I mean how the hell do you get a bunch of amino acids to assemble themselves into an functioning eye, leg or wing.  Well it turns out we can do that.

And social formations are the same. Not only must information be effectively generated, handled and transmitted, but there is the further complication of people’s interests.  Economics usually collapses these considerations into questions of interests.  It sees the problems as ‘principal/agent’ problems and they’re solved – to the extent that they can be solved – by mechanism design – by rules – like the rule limiting each side’s appeals or ignoring outlying votes when judges vote on who did the best dive.

Very few worldly situations are fully ‘incentive compatible’. Indeed that’s usually true even for the more rarefied situations that turn up in formal models. Still there’s another string to our bow.  We also lean heavily on the psychology and ethical intuitions that evolved with the evolution of our species. The fact is that we evolved as a closely cooperating species with a very strong sense of group solidarity.

So we play on that. Soldiers march, sportspeople train together and business and public sector agencies bond in a thousand ways – they go on retreats and discuss their organisations ‘vision’ and ‘mission’ and they may forget the mission and the vision on Monday morning, but they all feel more together (Well many of them do.) And our headmaster and the church and the St James Ethics centre all preach that we should do the Right Thing and on and on it goes. In any event, if we couldn’t get people to buy into the mission of their agencies, their families, their choirs, their firms, their football teams, their nations then we’d still pretty much be hanging from the trees. (OK we never hung from trees, but you get my drift.)

Now I’d argue that the whole story of development is actually the story of growing complexity. That seems like a reasonable description of the evolution of our universe which started off as something stranger than I can imagine, but spent a lot of the next few eons as mostly hydrogen. It’s true of the evolution of life – you know from jellyfish to mammals and thence to smart phone enabled humangoes. Harvard’s Ricardo Hausmann and César A. Hidalgo think that the story of economic development is the story of the evolution of economic complexity. But if development is the evolution of complexity, then, just as the jellyfish had a lot of work to do, a lot of jellyfish retreats to go on before it came up with a brain and eyes that could look around for threats and opportunities and make appropriate decisions, then one of the central tasks of development is expanding our mastery of the governance of complexity.

And the growth of complexity involves the mastery of detail – of finding out whether the captain really should challenge the LBW decision – or whether to do so would be a bad bet. Modularity is one means by which complexity is governed, wether we’re talking of inanimate matter (atoms, molecules) biology or social formations. So the jellyfish needs to develop a bunch of organs to handle different tasks which are handled by modules – stomachs, eyes, livers, kidneys, blood system etc. But all those tasks require coordination. Sometimes this can be done using simple (brainless) thermostat like systems. And sometimes one needs something more complex in handling the interactions than that – a brain for instance. But the brain is a ‘top down’ coordinator. It has its uses in mobilising large amounts of resources, but it will have trouble in handling the micro-detail. That needs to be delegated to sub-systems – the modules.

Adam Smith’s story of the growth of the market and the division of labour is a similar story – though it’s without any brain. He’s not silly enough to think that the system can do without a brain – the brain is the government and the laws and mores of the land – those things that permit co-ordinated action. But he’s a fan of the power of decentralised decision making.

This story of the governance of complexity is also a good way to tell the story of the evolution of modern government. Separation of powers between ‘modules’ or routines of government helped tame the evils of excess power. But governments have grown since the eighteenth century when they came up with these ideas. You can complain about big government, but the fact is that governments have become immensely more complex – mirroring the growing complexity of the market and society. Only a fanatic or ideologue would imagine it should be any other way. (This is not an argument for larger or smaller government than we have, though it is an argument for a much more complex, capable government than we had in the eighteenth century.) The ‘web of law’ weaved by governments is immensely complex with thousands of organisations each having to interact according to law.  And there’s a very imperfect ‘brain’ at the top which gets involved when absolutely necessary, but even while people go on about the holy grail of ‘joined up government’ or ‘whole of government’ delivery of services the fact is that the central brain is a bit of a last resort. Certainly not much chop at the micro-detail.

Those who have been following my posts about regulation will see the relevance of this thinking in my mind to regulation. I’m arguing that the problem with regulation is that most of the high level stuff doesn’t cut it. That regulation involves finessing the micro-detail – just as running a car factory does – or a bank, or an eye or a liver or a circulatory system or pretty much any damn thing.  In business the story of decentralisation is critical. Small businesses make decentralised decisions by definition and large businesses must consider how to decentralise decision making. Nineteenth century railroad companies had to figure out how to do it, and Toyota figured out how to do this better with making cars, and they’re still not very good at it with banking and the incentives are all skewed so it’s not where the best bankers’ headspace is at.

But one of the great corruptions of our ability to handle complexity is command and control – the appeal to the central brain. In politics when the British Parliament tried to curtail the King’s power and bring it more firmly within the web of law, there was carnage for 20 years between 1640 and 1660. Then just 40 years later things fell into place.  A fair bit of the thinking had been done in the civil war between 1640-60 and some lessons were learned.  A Parliament knew pretty much what it wanted – and lo and behold  William of Orange had been good enough to marry Mary Stewart in Holland, so they could be imported. A way was found to overcome ‘command and control’, to push authority down to the next level and voila – Britain powered into the modern era. The new constitution was better – much better – at governing complexity.

Finally then, at least for this post, this gives us a useful way of thinking about an agenda for the future. It’s the root of my own dislike of populism. In lots of areas of both commerce and government we have lots of trouble getting away from the limitations of ‘top down’. The civil war example is instructive, because the answer to the limitations of top down management is not to simply throw the switch to decentralisation. Centralisation is a crucial routine, indeed a module, even if it’s a meta-module. One very often needs some capacity at the top to arbitrate between different interests or focuses in the organism. So simply flattening an organisation or making it more democratic may be like sticking eyes and a liver into a jellyfish. They won’t help because they’re not properly evolved as part of the whole. The transmission of power to the parliament in the Glorious Revolution of 1688 required a lot of things to fall into place. Together.

Rather we need to apply ourselves to thinking of the obstacles to greater delegation from the top, and we need to build the institutions to do so. So when I argue for independent fiscal policy advisor that’s because I know that handling it in the way we do dumbs it down. We need decisions on fiscal policy made in as informed a way as possible and an important way of doing so is building an institution which enriches the decision which is ultimately sanctioned by the democratic process.  That’s what we do with monetary policy and it works a lot better.  When the US Fed decided on new and stronger “quantitative easing” measures a few months back, Mitt Romney said it was a travesty. The fact that it had the support of all but one of the twelve major decision makers in the Fed Reserve System was of no significance. Mitt knew, along with Ron Paul and all the other Republican know-nothing fanatics. My point is that their penchant for ‘top down’ direction by a body that didn’t know illustrates poor governance of complexity – with the mistake being the classic mistake of ‘top down’ interference.

When, as Ken documented for us, Michael Harmer abuses the processes of our law to aid a political agenda, the electorate – the ultimate ‘top down’ authority and central brain in our political system – is ill equipped to consider its merits or otherwise. That’s why such matters are rightly delegated to courts – though behind courts also lies the institution of the legal profession, law societies etc. (It’s interesting to note here that making something more ‘democratic’ sounds like you’re decentralising it – but in this case we’re moving further up the hierarchy of decision making – to the top where domain knowledge is scarce.  In this case institutions that we have already developed – which help decentralise decision making (though not without self-serving rent acquisition by lawyers) are being wound back. That’s a backward step in the evolution of complexity and our capacity to govern it.

Though I have little doubt that we’re continuing on our longer journey towards greater decentralisation and capacity to govern complexity.

This entry was posted in Economics and public policy, History, Life, Philosophy, Political theory. Bookmark the permalink.

12 Responses to Governance

  1. Mike Pepperday says:

    Just to clarify: you are suggesting independence like the RBA or the Electoral Commission – yes? And the independence would be achieved by having the fiscal commissioner being appointed by parliament and reporting to parliament, not the executive?

  2. Nicholas Gruen says:

    Hi Mike,

    For the purposes of this discussion I’m not sure the distinction is that important. Thinking about it, isn’t the standard procedure with ‘independent’ appointments like that to the ABC or the RBA that the officers are selected by the executive, but responsible to the parliament through their enabling acts?

    The point is that this structure – even if the appointment and reporting arrangements were different – is still more decentralised than having such things handled by civil servants who are conceived of as simply executing the will of their political masters (though through the auspices of the civil service which introduces some additional legal structure to the situation). By contrast an appointee operating under a specific act or some well specified independence then has a much greater level of independence, implying a greater degree of delegation and decentralisation of power.

  3. Stephen Bounds says:

    Hi Nicholas,

    Interesting article – although I think you are clouding the picture by talking about democracy. I think it’s more appropriate to talk about the method of determining decision-makers, ie is it:

    • dictated – the right to make decisions is asserted or claimed
    • delegated – the right to make decisions is granted by all to a few
    • franchised – everyone who fits some class definition has responsibility to make decisions

    The other aspect is the basis for asserting quality of the decision-makers, for example, technical ability, divine right, representativeness etc.

    You aren’t really advocating a switch away from command and control, but a form of technocratic decision-making through delegation. The opposite of command and control is distributed problem solving through independent systems, ie the ability for state governments to set education standards.

    Also I must say, it appears that we are becoming much more command-and-control oriented through the leverage of financial purse-strings. Was your last observation meant to reflect long-term trends or the current situation?

    • Tel says:

      Also I must say, it appears that we are becoming much more command-and-control oriented …

      Shhh… Don’t mention the socialism.

      I slipped up a few times, but I think I got away with it.

      You are right though, Nick tends to presume that information must flow to the centre and then back out again. When considering ways to fix the Principle-Agent problem, never overlook the obvious solution … get rid of the Agent!

      • Nicholas Gruen says:

        Thanks Tel,

        My examples might suggest that “Nick tends to presume that information must flow to the centre and then back out again”, but that’s just because the examples aren’t completely representative of the way I’m thinking. But I don’t think that. In fact if you read the piece carefully (though I also admit it could have been written better) you’ll find both examples and explicit statements that suggest otherwise.

        So the Toyota Production System which I cited does get rid of agents – it doesn’t require all the information back at the centre.

        And then there’s this statement.

        But all those tasks require coordination. Sometimes this can be done using simple (brainless) thermostat like systems. And sometimes one needs something more complex in handling the interactions than that – a brain for instance. But the brain is a ’top down’ coordinator. It has its uses in mobilising large amounts of resources, but it will have trouble in handling the micro-detail. That needs to be delegated to sub-systems – the modules.

        • Tel says:

          My apologies, I was a bit glib and off the cuff. There’s a lot more nuance to what you were saying.

          I’ll try to be a bit more nuanced myself. Just because government has become big and complex and centralized doesn’t imply there’s a genuine need for us all to be constantly governed by such an entity. All it proves is that government has been capable of grabbing that power on the excuse of fixing problems (often problems earlier governments created) and thus convincing people that they are helpless and in need of supervision.

          You use Ron Paul’s dislike of QE as an example of top-down decision making (Mitt Romney probably is a top-down kind of guy so fair example), as if somehow Ron Paul thinks he knows how to manage the Fed differently. What Ron Paul has been saying is that if he were running the Fed, his first task would be to shake everyone’s hand and send them home, then put a hammer through the printing press and and a “for sale” sign on the building. Ron Paul doesn’t believe there should be a Fed at all! People are perfectly capable of organizing accounts between themselves if they are permitted the freedom to do so. If a bank goes bust, there’s a reason for that so get it done, and get it sorted, and get on with business.

          I guess what you are saying is that central government should beneficently allow some of its power to carefully devolve to more localized government entities, but what Ron Paul is saying is that certain powers (such as trade and currency) evolve up from the people and never should have been snatched by government in the first place, and over time we will just have to get enough people determined to take those powers back. Quite a different way of seeing things.

          By the way, I don’t believe for a moment that the Federal Reserve is anywhere near as politically independent as the RBA. I also think that sooner or later this QE? just has to wash down into inflation. Their approach is a short-term fix at best, I doubt it will even last out the four years of Obama’s current term before the inflation shows, and once that momentum starts, I have no idea how they intend to bring it back under control.

          Let me also point out that while Keynesians wring hands over the Tragedy of Thrift, if you allow the natural emergence of currency from the bottom up, it sorts itself automatically as hoarding of any particular item just results in people moving across to using a more convenient item as currency (whatever that may be) and this devalues the hoarded items just sufficiently to bring them back into service. There’s no particular disaster in having multiple currencies in circulation, any more than we have multiple options for investment or saving.

  4. Nicholas Gruen says:

    Thanks Stephen,

    You’re right – the post runs together two things that ultimately need to be considered, and which are not wholly distinct, but which should have been introduced quite separately. One is the governance of productive work where the central aim is expediency. The other is political governance of a community where the considerations are much broader.

    Certainly in the examples I’ve given of institutional structures of government, I can see why you’ve called what I’ve described ‘technocratic’. But that’s just the examples. ‘Technocracy’ seems to me to have some element of top down choosing – with those in authority doing the choosing, but that’s not necessary.

    That’s ultimately extraneous to my argument – which is about evolved forms of decentralisation – however they evolve and however the decentralisation occurs.

    I agree with you on short term trends. There are quite a few things that seem to be becoming more command and control. I was thinking of the longer term tendencies.

    One observation your comments on this lead me to is that we’re suckers for an artificial market. I guess it makes us think of the virtues of decentralised decision making a la real markets. But the ‘market’ in academic excellence which gets down to how many articles you can publish in high prestige journals is a kind of faux decentralisation which can end up being worse than what came before – involving more poorly chosen criteria for success than obtained previously.

    • Stephen Bounds says:

      Hi Nicholas,

      I’m using technocracy in the sense that “experts are the ones who should make the decisions”. It doesn’t have to be a top-down appointment, I would see a society that accepts the concept of only giving people with a masters in economics a vote as also “technocratic”. Structures also matter; the USA system, with an appointed executive, is far more technocratic than the Australian system where Ministers must be drawn from the ranks of elected parliamentarians.

      I’m not sure if you’ve come across Dave Snowden before? He has defined three heuristics of effective complexity management:

      • Use finely grained objects (two aspects: individual stories matter and shouldn’t simply be abstracted away, and organisations aren’t uniform but consist of lots of independent or quasi-independent actors)
      • Leverage distributed cognition (but not necessarily distributed decision making)
      • Disintermediate as much as possible (removing mediating layers between decision makers and raw data)

      A true marketplace works because all of these tests are met. And the vast majority of government policy completely ignores them. I’m particularly disappointed in the indigenous intervention programs, for example, because they impose command-and-control decisionmaking into an extraordinarily complex space which should be devolved.

      With your specific example on fiscal policy, the appointment isn’t sufficient. Even with an independent body, if they only made their decision based upon aggregate statistics on Australian economic performance, for example, then they would lack the ability to effectively identify and respond to negative outcomes that were masked by the abstraction.

  5. Tel says:

    So simply flattening an organisation or making it more democratic may be like sticking eyes and a liver into a jellyfish. They won’t help because they’re not properly evolved as part of the whole.

    It’s a fair point, but also a trade off. The advantage of a Principle-Agent structure is division of labour (often considered to have efficiency gains), the disadvantage is centralization of trust and the creation of rigidity and points of failure. Ultimately, you can never really outsource trust. You can outsource any physical activity, or even outsource the tedious research and handle cranking of a calculation, but once you trust the guy blindly, you are screwed.

    When the biggest and meanest ocean predator is drift nets, the jellyfish start to look pretty smart.

  6. Nicholas Gruen says:

    Thanks Tel,

    This is a response to you above timestamped [December 22, 2012 at 5:11 am] – you keep interesting hours – or perhaps are overseas – half your luck.

    On Ron Paul, yes, in a sense a bad example for me to mention because Paul does seem to have principled objections to QE based on his belief in an antediluvian monetary system. Crazy is as crazy does. Mitt has no such excuse. Just a bit of know all know nothingness from the guy who wants to be on top.

    A better example would have been anti-dumping policy. Here you have a case where it can be explained to anyone who’s prepared to concentrate for a minute or so why imposing special anti-dumping duties is not a good idea. It’s not like it’s a tradeoff between capital and labour or anything – it’s just a bad idea. BHP-Billiton’s successful anti-dumping action saves a few jobs in steel and destroys more jobs in metal manufacturing and further downstream.

    The Productivity Commission says this in its report, but the government has to proceed with policy that can be explained in a sound bite on the radio – which is to say they’re ‘strengthening’ the existing anti-dumping policy. That’s an example of the kind of thing I’m talking about – a failure of governance. This, by the way, is a kind of meta-political criticism of the Government. In this case the Government is afraid of democracy itself – of the ill informed ‘vox pop‘ kind. This shows that ‘democracy’ is not of itself necessarily always the appropriate form of delegation of decision making.

    I don’t really think of this as a technocratic point, because I have a fair bit of faith that if one put together some kind of citizen deliberation mechanism – a citizen’s jury if you will – it would come to the right conclusion on this subject (and I don’t really want to impose my idea of the right solution if the electorate can’t be persuaded – if not en masse then via the right vehicle (with the citizens’ jury being a special purpose cognitive elite).

  7. Mike Pepperday says:

    Tel would have it that doing away with the reserve bank would leave people free to do as they wish. But it wouldn’t: it would simply be ramming his (or Ron Paul’s) notion down everyone’s throat. There is only one way to find out what the people wish and that is to see what they actually do if they are in charge. My bet (reason below) would be that they would stick with the RBA.

    And Nick would have it that making something like that democratic would be like giving a jellyfish eyes—it wouldn’t know what to do. Nick, you’re against populism but, like prostitution, no one is really for it. No one suggests it. I think you’re using “populism” as a pejorative for “democracy,” i.e., as a dismissive put-down of democracy. Prostitution is not the same as sex.

    A lot of this talk is surreal to me. I ask myself, why guess? Why not take a look and see? Is there any example of democracy (not populism) being like eyes on a jellyfish? I’d genuinely like to hear of it. The thing is, an “independent” commission has to be answerable to someone and if, as is often the case, it answers to parliament then that is actually more democratic than answering to the executive because it gives the people more say in ruling. I don’t know of any systematic studies of this phenomenon of independent bodies. If someone does, I’d like to know of that, too.

    Why speculate in the abstract? The admonishing example is Switzerland where the people rule to the extent where, it is said (with some exaggeration), that S has no government, only an administration. Worried about “coordination”? No problem: leave it to the people. S sits there, smack in the middle of crumbling Europe, staring out at us, silently saying, “WTF do all you other countries think you are doing?”

    In Switzerland there is no question of Big Capital telling the government what level of mining tax it may levy, or dictating what gambling regulation it will tolerate. Yet S is the very place where capital flees when the governments of the rest of Europe make a hash of it. In S it seems they never make a hash of it. In the Swiss context it is laughable to talk of a paltry “independence” of a bank or broadcaster or electoral commission. No bill—not one, on anything—becomes law without the people’s approval, implicit or explicit (the people knock back 3% of bills). No executive was ever worried about being kicked out for its immigration policy (which is a big issue in S); no executive ever got kicked out for anything. Ever. With very rare exceptions ministers of state retire when they have had enough. It is incredible the nonsense the rest of the so-called democratic world indulges in with its desperately-fought elections.

    History, since 1215 is quite clear: the more democracy the better. Yet the King Johns never tire of saying the ordinary barons can’t be trusted to govern. But just look at Switzerland. What on earth is the worry about democracy? Which is the best governed country in the world? S is high in the short list. Which country has the best record of governance for the last century and a half? S by a vast, vast margin. There they are, lederhosen-clad, yodelling yokels running around the hills governing themselves and showing the governance of almost the whole of the rest of the world to be incompetent.

    I don’t know but I expect, Tel, that S has a federal bank doing its thing, as here and as in the US. Since nobody has yet convinced the people of Switzerland to abandon it that alone tells me—a person with only a hazy idea of what monetary and fiscal mean—that a federal reserve bank is a good idea.

    Of course, it had better be independent.

  8. Nicholas Gruen says:


    I wasn’t ‘speculating in the abstract’. I wasn’t trying to establish ahistorical rules. I was giving examples. A lot of what you say I agree with. In fact pretty much all of it except where it disagrees with me where I think it’s just misunderstanding what I’m trying to say. You seem to be wanting me to be saying things that you disagree with. Why don’t you try a different tack? Why don’t you read my words as saying things that are close to what you think make sense. Then we might be able to enlighten each other from our different perspectives.

Leave a Reply

Your email address will not be published. Required fields are marked *

Notify me of followup comments via e-mail. You can also subscribe without commenting.