We are developing the social individualist meta-context for the future. From the very serious to the extremely frivolous... lets see what is on the mind of the Samizdata people.

Samizdata, derived from Samizdat /n. - a system of clandestine publication of banned literature in the USSR [Russ.,= self-publishing house]

Tim Harford on the power of bottom-up decision making (and on H. R. McMaster)

If you haven’t already partaken of this bit of video, then you really should. It lasts just under twenty minutes.

Tim Harford is speaking, in 2011, at some gathering of the clever and the smug, but it’s better than that. The name H. R. McMaster comes up several times, and this is, among other things, a very good quick way to learn why McMaster’s appointment by President Trump as his National Security Adviser might turn out to be such a very good one. It certainly explains why this appointment is already so very popular. You don’t have to believe that the USA rearranging matters in faraway countries is always or even ever a wise policy to get the points that Harford is making.

Harford also mentions, in passing, Hayek. From this, you may guess that this is a talk about decentralised decision making, and how on the spot knowledge, again and again, trumps the wisdom of the Central Committee or the High Command. If that is your guess, you would not be wrong.

The story that Harford tells reminds me of another transformation of policy that happened in China, and gave rise to what is now called the Chinese economic miracle. This miracle is now starting to look rather less miraculous, but it was still a massive improvement over what preceded it. That change too is usually attributed to a change of top leadership and of its top-down policies, but that policy also, I seem to recall reading, began at the bottom of the chain of command and in spite of the chain of command.

I even seem to recall having linked to stuff about that from here. Yes, here is that posting, about teeth of all things, and here is the article at Planet Money that the posting linked to. It’s the same story as Harford tells in the above-linked-to video.

Tweet about this on TwitterShare on FacebookShare on TumblrShare on RedditShare on Google+Share on VKEmail this to someone

17 comments to Tim Harford on the power of bottom-up decision making (and on H. R. McMaster)

  • Paul Marks

    Interesting.

  • Brian cites Tim Harford of around 6 years ago who starts with mistakes made 12 years ago. [And I did go view the whole 18+ minutes – though that was a close-run decision after about 3 of them.]

    The theme is (I hope I have it right) that central control is often wrong; that sensible decisions made on the ground are often better. I believe instinctively that this is right: there should be a lot of autonomy for those doing the job (though clearly within true high-level objectives set by the higher up).

    But I must ask, was not the decision to commit the Haditha massacre made there on the ground? Surely Donald Rumsfeld did not issue terms of engagement calling for the killing of non-combatants (including women and children). This makes me think the obvious: that serious mistakes are made within pretty much all command structures and within pretty much all terms of engagement.

    Though I believe in a policy of on-the-ground autonomy (of how if not what), I don’t see that this particular narrative is evidentially anywhere conclusive.

    Finally, and most importantly, I note the publication date of (Brian’s link to) quotulatiousness.ca: it is yesterday 22 February 2017. Given the talk is most approving of Lt Gen H R McMaster in previous roles, and he has just been nominated as USA National Security Advisor, I must wonder whether the motivation of quotulatiousness.ca was totally driven by reiterating the desirability of bottom-up decision making. [Not that I have anything against said general, nor even any prior knowledge of him.]

    Best regards

  • Hi Nigel. I can assure you that my interest in posting Tim Harford’s video in the first place was that it seemed quite topical (in light of the McMaster nomination for US National Security Advisor), and was called to my attention by Mr Harford re-linking to it on his Google+ account.

  • bobby b

    Top-down and bottom-up are both, in my experience, unfortunate situations brought about because tops aren’t interfacing with bottoms. They’re ways of working around the fact that your org is crippled.

    When you’re the top, if you aren’t constantly in touch with the people at the sharp end of the stick learning what’s working and what’s not working, you might as well be blind. Pushing for on-the-ground autonomy is often just a workaround for “my boss is an idiot, can’t we do without him?” Better to just have a good boss.

  • Julie near Chicago

    Yes, very interesting.

  • Itellyounothing

    Bobby B,

    There will always be a bad boss who needs mitigation, just as the best boss always listens to the front line, but not everyone gets the best boss. Error correction at lower levels is a great feature of ground up organisations, ideally with new ideas reaching the top via a competent top down leadership.

    The great thing about ground up is that from the top you can compare results and have evidence of the winner.

  • Lee Moore

    Hmm. It’s a good talk, but…

    …I think it’s confusing strategy and tactics. Strategy is what you do before you make contact with the enemy. Tactics is what you do after that. No plan survives contact with the enemy, so your tactics had better be flexible. Which is not to say that you should never change a losing strategy, but strategy has more inertia. You can’t change it often without ceasing to have a strategy at all. Tactics, by their nature, benefit from useful feedback from the enemy (or market if we’re talking business) and so are readily improved on the spot. It’s a bit like engineering – you try it and see if it works. If not, you twiddle it. Strategy, by its nature, is usually concocted in the absence of feedback from the enemy. It’s more like philosophy. You can take a view on whether an argument looks good on paper, but there’s no real test until you move on to implementation.

    Rumsfeld, who figures less heroically in the talk than HR McMaster, was essentially a strategist. His thing was about redesigning the US military to deal with the post Cold War world. He had no particular expertise in actually running wars. So when he got to run one, it’s not particularly surprising that he didn’t perform very well. (Though, like the 1st Gulf War, the crash, bang, wallop phase of the 2nd Gulf War, which must have been devised by HQ staff not by soldiers on the ground, was spectacularly successful.)

    HR McMaster, meanwhile, comes to a largely strategic role – advising the President on threats and what to do about them – from a tactical background. He seems a clever chap, and may well adapt well. And tactical expertise => strategic expertise is the natural progression. Even Alexander started off skirmishing before he moved on to conquering the world.

    Overall I have a certain sympathy for Nigel Sedgwick’s view. You can have good and bad top dogs, and good and bad bottom dogs. Generally speaking a bad top dog will do you a lot more harm than a bad bottom dog. But Harford is certainly right that if your top dogs’ strategy does not allow for tactical flexibility by the bottom dogs, you’re doomed.

  • A little something on another general from a few months back.

    http://www.samizdata.net/2016/11/from-al-ghazali-to-general-douglas-haig/

    As to good and bad top dogs in Trump’s current line up? Jeff Sessions is in the wrong position. He is already getting push back on civil asset forfeiture from his own party. A topic for another day.

  • Mr Ed

    I am reporting the post above this (as I write) to the Security Council of the United Nations for a blatant breach of the Haig Convention.

  • Mr Black

    I don’t agree with the basic premise here, that lower level decision making is superior in and of itself. People at a lower level have (generally) less understanding, less education and quite often less intelligence than the people above them. Listening to the ideas of the people at the coal face will result in a huge pile of garbage ideas and if you’re lucky, one or two great ones. The ideas themselves are almost beside the point when it comes to changing an organisation. You need 3 types of courage to turn an idea into an action. A subordinate with the courage to try something outside the established norms, the courage of a superior to allow people to put their new ideas into practice (and possibly fail) and the courage as top leadership to adopt ideas that are not their own, when they have been proven to work.

    Almost all these types of stories are not about local knowledge solving problems but capture the workings of a genius while they were still in a junior position, but none the less, still a genius. They have amazing problem solving ability at every level and walking back through their careers you will see it time and time again. Rommel made his name inside the military leading an infantry company. He really made his name leading a panzer army in Africa. Same genius doing both. What they need more than anything is a superior above them who recognises that genius and does not try to restrain it. McMaster could have been sidelined into a logistics or personnel role if he’d had a less forgiving officer above him.

    But I guarantee that all the other Colonels had their own ideas too, but they didn’t have the genius and courage to push ahead with them. What we have here is someone looking at the diamond in the coal and forgetting about all the coal.

  • Alisa

    People at a lower level have (generally) less understanding, less education and quite often less intelligence than the people above them.

    I disagree (generally). Not that I think that those at the lower level are smarter (in whatever sense), I just think that generally, there is not necessarily a significant gap in those qualities between the top and the bottom. More often than not, people at the top are there by virtue of better “climbing” skills, with all which that implies, for better or worse (I am not implying anything negative here). Ideally, they also have more experience than those at the bottom, having spent enough time at the bottom themselves, at the beginning and through their climbing process. They are also positioned to have the ‘bigger picture’ – in part, by virtue of sitting at the top. But they can’t really have that without constant input from the bottom, including the kind of input they may perceive as contradicting their take on things. To ensure that such input is of real value, the people at the bottom should be about as intelligent as those at the top. And ideally, the whole thing should be a two-way street.

  • Mr Black (February 25, 2017 at 6:29 am): “Listening to the ideas of the people at the coal face will result in a huge pile of garbage ideas and if you’re lucky, one or two great ones.”

    When the ideas are about how we can mine coal fast and safely, listening to the ideas of the people not at the coal face, who get their ideas by imagining what the coal face might be like (because they have never themselves been anywhere near it), also produces many ideas, also containing much folly. The point is not that any group has all, or even mostly, good ideas. It is that an awful idea will be seen to be awful by those at the coal face when they experience it, whereas those who merely imagined what life at the coal face was like will impose their ‘brilliant’ idea and then imagine that it’s working fine. The rare good ideas that either group will produce will be less easily recognised.

    “People at a lower level have (generally) less understanding, less education and quite often less intelligence than the people above them.”

    They will usually have more understanding of coal mining. Intelligence has many aspects: “Professors of physics do not necessarily make good baseball players.” (Thomas Sowell). If education is both offered disproportionately on political/social grounds and marked on political/social grounds, less of it may sometimes be more: real life educates most of all.

  • Lee Moore

    there is not necessarily a significant gap in those qualities between the top and the bottom. More often than not, people at the top are there by virtue of better “climbing” skills

    It is true that different skills are not perfectly correlated with each other, so that the most skilled “climber” may not be the most skilled marksman or the most skilled troop leader. But it is equally true that intellectual skills are correlated to some extent – a non trivial extent. So it would actually be very surprising if the superior “climbing” skills of top dogs weren’t also statistically predictive of superior other skills – including ones actually relevant for actually performing well in higher level roles. Those at the top really are smarter than those at the bottom.

    At best, it’s fair to say that someone of reasonably high general smartness – we’ll call it an IQ of 120 for the sake of calling it something – is likely to rise higher in an organisation if his climbing skills correlate with the average of 130 IQ folk, and his mapreading skills are more 110 ish; than if his ability mix was the other way round, 110 climbing and 130 mapreading.

  • PersonFromPorlock

    Mr Black
    February 25, 2017 at 6:29 am

    People at a lower level have (generally) less understanding, less education and quite often less intelligence than the people above them.

    It’s an aside, but I recall a study the US Air Force did back in the late ‘sixties which found that there was a negative correlation between intelligence and officer grade. I also recall that the Air Force’s response to it was that senior officers were still ‘smart enough’, which made me wonder at the time if they really were.

  • Julie near Chicago

    Niall quotes Dr. Sowell above:

    ‘Intelligence has many aspects: “Professors of physics do not necessarily make good baseball players.”’

    Really good businessmen (CEOs, presidents, founders of businesses who make great successes of them) are not famous for their smarts as economists. And as far as I know, few economists, including the really good ones, are on the list of the world’s top thousand (or whatever) business successes.

    Also, speaking of the Air Force, I suppose everybody has heard that the Powers That Were during WW II sought pilots who were not terrifically imaginative, because the imaginative ones were more likely to panic in the face of disaster and either freeze or become incapable of making a plan (on the very quick fly!) and then carrying through on it.

  • David Moore

    Mr Black

    I don’t agree with the basic premise here, that lower level decision making is superior in and of itself. People at a lower level have (generally) less understanding, less education and quite often less intelligence than the people above them.

    Generally, and in particular, in the US Army, I don’t believe this is in anyway true. Just look at the process by which you achieve advancement to the higher levels;

    http://www.johntreed.com/blogs/john-t-reed-s-blog-about-military-matters/60879683-the-u-s-military-s-marathon-30-year-single-elimination-suck-up-tournament-or-how-america-selects-its-generals