The story so far. In our last exciting installment, we argued that there are three fundamental aspects of economic life that prosper in markets are
1. the pursuit of self interest
2. the generation and utilisation of information and knowledge throughout the economy, not just at some executive centre
3. the joint production of value by feedback between economic agents and the learning it produces.I suggested that while the first two aspects are at the centre of the neoclassical and of Hayek’s ideas of the economy (respectively), the importance of the third aspect tends to be underplayed.
In this installment I argue that feedback and interactivity in learning has progressively grown in importance and continues to do so. Here the focus is on the workplace, but a subsequent post will venture further afield. I also suggest an interpretation of modern management methods that is an extension of Hayek’s idea that engineering knowledge must be complemented by knowledge which is more localised.
Adam Smith tells us that slavery is the most expensive form of production when it seems to be the least expensive. Why? Because the abusive relationship between master and slave is one of punitive rather than productive feedback. The slave won’t propose improvements in the way he works to his master because the master will capture all the gains realised. Indeed, the slave may fear that he’ll qualify for further abuse either for not having pointed out the problem before or for being lazy in proposing some way of reducing his work-load.
Smith also looks at other forms of relationship including various kinds of land tenure stressing the way in which behaviour is influenced by incentives. He’s fairly jaundiced about employers he nominated the apprenticeship system which saw employers combine against employees as the worst abuse of the mercantile system. He concludes that labour responds well to encouragement of all kinds including pecuniary.
The liberal reward of labour . . . increases the industry of the common people. The wages of labour are the encouragement of industry, which, like every other human quality, improves in proportion to the encouragement it receives. A plentiful subsistence increases the bodily strength of the labourer, and the comfortable hope of bettering his condition, and of ending his days, perhaps, in ease and plenty, animates him to exert that strength to the utmost.
But what of feedback on the production process in a large factory? Certainly large ‘Taylorist’ establishments in the early 20th century used methods like suggestion boxes to elicit productive suggestions and some went as far as prizes and other rewards for good suggestions. Piece rates were also a popular way to ‘incentivise’ workers to optimise their productivity. But at least compared with modern ‘Post-Taylorist’ management, these methods worked poorly.
The big problems are social and motivational. Monetary rewards can undermine fellow feeling between employers and so their co-operation as employees question the justice of different decisions and jockey for advantage with one another and/or collaborate to snow management. And it’s very difficult to measure a single labourer’s productivity within a highly interdependent centrally designed and directed productive system.
But from the 1950s on the Japanese under the influence of an American statistician Edwards Deming worked out a new system of production the central purpose of which was to elicit the active collaboration of the workforce, not just in the design, but in the endless optimisation of the production process. The system seeks to maximise feedback between the different productive units within a large productive system and indeed between that productive system and those that interface with it (particularly amongst suppliers and customers).
Before elaborating further, let’s segue back to our friend Friedrich Hayek. In his mid 1940s essays collected in Scientism and the Study of Society Hayek wrote of the importance of not privileging the universalist ‘scientific’ knowledge of the scientist or engineer over “the knowledge of the particular circumstances of time and place” of the trader. (This puts him well ahead of his time compared with Schumpeter and Veblin in the generation preceding him both of whom anticipated some ultimate triumph of the engineers.)
Compared with the work of the engineer that of the merchant . . . contributes a step towards the achievement now of one end, now of another and hardly ever is concerned with the complete process that serves a final need. What concerns him is not the achievement of a particular final result of the complete process in which he takes part, but the best use of the particular means of which he knows. . . . 1hough 2 knowledge is not of a kind which can be formulated in generic propositions or acquired once and for all, and though in an age of Science it is for that reason regarded as knowledge of an inferior kind, it is for all practical purposes no less important than scientific knowledge.
In a way that is strikingly similar to Hayek’s insistence on the complementary importance of traders and engineers’ modes of knowledge within an economy, the Japanese production methods likewise insisted on the importance of the knowledge of the shop floor complementing the engineering knowledge of the managers in a large manufacturer.
Where traders are great integrators of local knowledge in a large market, in the extended and centrally planned order of a large multinational manufacturer, Total Quality Management (TQM) * was one of the first conscious attempts to harness the value of information throughout a whole production system (rather than conceptualise it as residing exclusively amongst the managers and engineers). TQM single-mindedly focuses on maximising feedback between one employee and another, between workers and management, between one group or division and another. There’s also an enthusiasm for capturing information outside the firm, most particularly from suppliers and customers. Today the instruction given employees and also suppliers is more open, allowing them to contribute their own unique knowledge.
Under the old system managers and engineers specified (sometimes in quite fine detail) the tasks for employees, and for suppliers. Once General Motors would have designed and minutely specified the brake it wanted and then tendered out the job and taken the lowest couple of bidders to jointly supply (to limit the scope a single supplier might have to behave monopolistically). Today Japanese firms specify much broader design parameters and then get a single trusted supplier (usually from an ‘extended family’ of regular suppliers often with equity sharing between the supplier and the customer) to assist in brake design. The potential benefits through this rich interaction and feedback are obvious.
Further, where most of Hayek’s writing on the importance of the information of the ‘trader’ has a fairly static quality,** the new system focuses almost obsessively not simply on the information the productive partners have (ie employees and suppliers), but on their capacity to endlessly optimise their own productivity and the joint productivity of the entire productive unit.
TQM is not just a few tricks a technique here or there. It involved extensive and interdependent changes of approach. As Western firms discovered to their cost, technocratic cherry picking from the system, produced all sorts of disasters. The architects of TQM saw themselves as building a system which was as much social as technical.
To maximise feedback and room for ‘lateral thinking’ there were conscious attempts to reduce hierarchical thinking (with accompanying flattening of management structures). But there’s nothing like getting rid of middle managers to re-discover what they were there for in the first place. Though it relinquished the control of minutiae from the centre, production decisions were made on the shop floor supported by extremely rich data collection on the production process, to help guide the exploratory process of continuous improvement.
Hayek describes the trader’s role vis- -vis the engineer or scientist as “in a sense much more “social”, i.e., interwoven with the free activities of other people.” Intriguingly while TQM is also more social consciously so than older production methods, where Hayek would (I think) see the trader as representing a kind of individualism, a critical aspect of TQM was to move away from individual approaches towards social collaboration.
TQM arranges the workforce into well trained, multi-skilled teams giving the production line great flexibility and capacity for innovation. Though teams undoubtedly have ‘negative’ benefits of providing surveillance against shirking, this appears to be a by-product of other aspects of the arrangements designed to improve morale and maximise co-operation.
In building teams and ‘quality circles’ as social groups the architects of TQM took advantage of the kinds of social norms that seem hard wired into our natures presumably from our evolutionary socio-biology. If contemporary theories I’ve discussed elsewhere are anything to go by, human consciousness and society emerged from the interactions both co-operative and Machiavellian of individuals in groups of primates (often of about the same size as TQM production teams!).
But TQM management goes to great lengths to encourage collaboration and undermine the more manipulative side of our natures. Thus employees are not ‘incentivised’ individually using piece rates or suggestion boxes with individual rewards. And one of Deming’s imperatives is to ‘drive out fear’. Fear of insecurity and individual penalties and rewards from management are seen as encouraging envy, resentment, buck-passing, and various other Machiavellian traits. Primatologists have argued that these kinds of Machiavellian manoeuvres may lie behind the evolution of human consciousness and society. But having done their work, the TQM managers try to keep them at bay to optimise the quality of collaboration and feedback within the system.
It all goes to illustrate ways in which Hayek’s insights about the importance of complements to the engineer’s scientific knowledge can be extended and the productive power of feedback in economic relations. Indeed, one could go so far as to argue that feedback is the ultimate economic category with price determination being a very specific sub-set of feedback between economic agents.
In our next exciting installment, we discover the neurological foundations of feedback between people or as Adam Smith called it ‘sympathy’. Yes folks, modern neuroscience really has uncovered the neuro-micro-foundations of Smith’s Theory of Moral Sentiments! There is no way to discover whether this is a silly exaggeration or a truth worthy of Troppodillians . . . . other than to read our next exciting installment coming to Troppo, sometime soon.
* The new approach has many names such as ‘post-Fordism’, ‘post-Taylorism’ and ‘just in time production’. There are no doubt various debates trying to specify differences between these terms. But I’ll use the expression total quality management or TQM to denote this new system.
** It was a surprise to me when I went looking for innovation in Scientism and the Study of Society that Hayek’s perspective on information was pretty static. He does write about the economy and society in an historical context, but he doesn’t really make innovation a focus of his discussion. This seems odd given that the dynamism of localised information adds another very powerful string to his anti central-planning bow. Perhaps the explanation is that Hayek spent his time developing his conception of information alongside and to a substantial extent in opposition to neoclassical economics and so his arguments are addressed to the ‘comparative static’ perspective of neoclassical economics. Another explanation is that I’m wrong it’s a while since I read Scientism and Hayek’s later work.
Good article. I remember thinking when I read through a six-sigma book that it is TQM with ninja rhetoric. The Agile Method also seeks to maximise feedback loops between developers and customers.
Yes, software is obviously a major item if one is considering the importance of feedback. So much so that some software now more or less writes itself on the strength of feedback alone – open source software. Users doing it for themselves. Watch this space!
Cam, I’ve just been to the link on Agile Software. I agree – the items on the right are important – just less important than those on the right. Feedback and DOING and letting systems grow out of that is more important than the systems (and software) which emerges from them.
This seems like a complicated way of saying ‘treat your employees as though you value and respect them (and their ideas), give them the chance to make a positive difference, emotional ‘payment’ as well as monetary and they will give you valuable feedback about what is actually happening on the production floor and who it could be bettered.
Treat them like imbeciles, or as if they be your mortal enemy, routinely sucking the lifeblood from your business and they will oblige with fulfulling your worst fears.
Employers can pay their staff and some pay them well, but they can never reimburse them for the chunks of time that work has subtracted from their lives–gone forever. A little emotional reward is the only recompense possible for the giving up one’s life. Money everywhere ebbs and flows, but time gets sucked and we are always in a state of atrophy.
A worker cannot take his pay packet or his savings beyond his deathbed but he can and should be able to take with him the knowledge and the feeling that his life’s labours were appreciated and well spent, lest he feel he has wasted his own life, by feathering the nest of another.
oooops sorry
‘. . . production floor and ‘how’ it could be bettered.
time gets sucked (away.
don’t it wot.
Very interesting Nicholas but (just to be a devil’s advicate) I think Link has a point.
Why is feedback in the workplace unique to market economies? If it improves productivity, an intelligent autocrat could demand that all his enterprise managers seek out and use “the knowledge of the shop floor” and appoint commissars to police the ruling. Market competition is a great discipline on managers but so is the fear of incarceration (self-interest).
Needless to say I am not advocating an autocracy (which would not be terribly cost-effective even if the ruler was a benign genius). But I really think the essence of market capitalism lies with the role of the price mechanism as generator of information across the economy rather than in the workplace.
Thanks Link and Fred,
Firstly Link I agree with you that the article is long winded! I should have edited it down more and made it punchier. I got a bit bogged down in editing it and could have improved it. But at some stage you have to press ‘publish’.
There’s also something else which I meant to add in this post, which slightly derailed discussion on the last installment. I really want to argue that feedback is a fundamental category of economics full stop. I also think that the significance of feedback to successful economic organisation is a powerful additional explanation for the superiority of markets over other structures in a whole bunch of economic activities (though obviously not all).
But the two arguments are separable, and my purpose is not to emphasise the second argument (which for instance Andrew Norton took me as doing in the first installment. Hayek’s anchoring so much of his own discussion about information in economics makes his own work less interesting and useful that it might otherwise have been in my opinion.
I think both of you underestimate what is involved in stimulating good useful feedback. Link says that it’s about treating people well. Well TQM tries to do that and that’s part of its essence, but in TQM treating people well is situated in a very specific set of practices. As I suggested in the post, the business world is littered with failures that tried to treat their employees well (Ansett said it thought of its employees as family). And it’s littered with poorly implemented TQM plans that made things worse. And TQM took a long time to emerge. So it’s not easy. Part of the point of this post is to argue that TQM is really a very highly developed form of organisation, not just a few tricks. I think of it as a stage in the ascent of human endeavour in which the blue collar workers came to be treated more like white collar workers and sought to intimate that by beginning the discussion with Adam Smith’s thoughts on slavery.
Another analogy
Oh Fred . . .,
I should have added that my point is not necessarily against autocracies but in favour of feedback whether one is operating in a market or within an autocracy (a firm or within the heirarchical structure of a government department). TQM systems are after all implemented within fundamentally autocratic structures. But those structures give a little in order to make themselves more productive.
So my point is not really pro market and anti-autocracy. It’s a case of horses for courses – markets will be better at some things and autocracies will be needed for other things. Rather I’m arguing that feedback is almost always good per se and secondly that, at least in autocracies it’s not that easy to optimise feedback. So we should be trying hard to do this and TQM gives us some very suggestive clues as to how we should be going about it. The place I’m thinking of most of course is governments. They’re into feedback like everyone else – one has to fight off the questionnaires from Govt agencies on what you thought of their service, but lets just say that often not a lot is done with that feedback.
Nicholas, Part of the point of this post is to argue that TQM is really a very highly developed form of organisation, not just a few tricks.
I agree. TQM is hugely important for globalisation. Because TQM/SPC/deming-way are so empirically rigourous in making a production process meet a quality output, it doesn’t matter where a product is built or fabricated. Something built in China or Mexico can meet the same quality requirements as something built in Germany or Japan.
I consider agile/Xtreme etc approaches to quality which maximise innovation and productivity without sacrificing customer expectations of quality.
Oh dear, another management fad which purports to ground the wage relationship firmly in nature herself. This time it seems, by the happy serendipity of natural selection itself fashioning neurones that act to facilitate the extraction of surplus value. Feedback a little understood or appreciated human attribute? Until recently, we didn’t appreciate the practical utility of telling each other what is going on eh? So the question becomes how can we encourage people to do more of the stuff that makes people human, in order to extract even more of the stuff that makes some people rich. Certainly it seems, some ungrateful wretches seem unable to appreciate their full obligation to harness every neurone to the service of their employers, so treating people nice just shows the limits of this kind of thing. Never mind, a few more sackings for ‘operatonal reasons’ and they will soon get the message.
“Tell us everything, even the things you might not think are useful, lest you be deemed recalcitrant and uncooperative.”
Give us a break. I need a smoke.
I don’t know why people decry the state of satire in the place. It is all around us.
Thank you Amused. I am, and especially liked
I can’t help but think the word ‘feedback’ could be replaced with ‘feedbag’, and the whole concept would still hold together and wouldn’t lose its meaning. Both are necessary in life, oftentimes compulsory and even impossible to avoid or do with out.
I think Ansett went belly-up not because they were nice to their staff, but because they got shafted by an even uglier beast,Qantas. Who are, by all accounts, horrible to their staff.
‘TQM’ has to be about human psychology and what’s considered valuable by humans. And that’s my point. I was making no criticism about your wording (well I was). No, they are good words, there are just too many of them and they state what I would have thought was obvious. But then I’m not an academic and like to get to the point quickly. The language used and concept such as ‘TQM’ risk obfuscating the issue. Not a good thing because the master/slave dynamic needs to realise it is over, lest the twain ne’er meet.(or agree) and they haven’t so far so I’m not holding my breath. Maybe your writing can help the idea of “the practical utility of telling each other what is going on . . .” more mainstream or summit?
The biggest issue I faced when I started working (and indeed at uni although I didn’t realise it) was mastering the subtleties of office feedback. What was fine in Rugby didn’t translate to the insecurities of modern professional services employees :)
Maybe a new industry for lateral economics?
Nick is, I think, talking about feedback in markets generally – between employer and employee, between a producer and their suppliers, between producer customer. He is talking about a broad process – a mindset, really – which has been refined to new peaks particularly by the car industry and then the software industry. He aims far beyond the let’s-create-quality-circles Western implementation of TQM whose popularity peaked in the 1980s, and which horrified people like Edwards Deming.
True TQM is far more than a fad. In the car industry kaizen and other elements of TQM have sent productivity and product quality soaring. (Car buffs who want to follow this further should pick up a copy of Womack and Roos’ The Machine That Changed The World.)
It is true that in theory some of the feedback processes described in The Machine That Changed The World could be implemented outside a market economy. In practice, it seems not to have happened very much at all. The market imposes a double pressure: it pushes you to elicit feedback (often uncomfortable) and then forces you to decide which feedback to act on. (Trying to act on every piece of feedback will slow down product development dramatically.) The market then ensures that those who elicit and act on the right feedback get greater rewards.
You might classify this as a second-round effect of self-interested knowledge utilisation. But it is definitely underplayed. The role of feedback effects has expanded dramatically over the past 60 years, and I suspect its expansion is actually accelerating as the Internet makes it easier to gather a much richer, more continuous stream of feedback.
(Note to self: Bone up on Hayek. Bone up even more on Smith, whose merits Nick continues to reveal.)
I’m teaching Quality Management at the moment as part of my summer sojourn in a Business School. Picking up on cam’s point, QM has progressed beyond TQM (and I once wrote a paper on management faddism so I’m well aware of the irony) and in fact feedback is key to work organisation and not just product development. There is certainly an implicit theory of knowledge embedded in QM which values autonomy and autopoeitic organisation rather than any form of command or control perspective (which of course has been by far the dominant form of organisation under capitalism). It is wise, however, not to get carried away on this path because just as aspects of human psychology can produce much greater innovation and creativity through autonomous organisation and continuous feedback, so too do the conflictual aspects continually work to undermine these processes.
I’d suggest as well as boning up on Hayek, people interested in these forms of organisation and knowledge utilisation bone up on the late German sociologist/philosopher Niklas Luhmann.
My two cents.
Of course, the bloke who most thoroughly worked out the importance of feedback in human thinking was Karl Popper. (Sorry, Nick.) The power of feedback underpins all of Popper’s work on political philosophy and philosophy of science. He would have understood The Machine That Changed The World completely. Indeed, you can see the Japanese-refined TQM process as the triumph of Popperian methodology, the power of continuous change and refinement, Popper’s “revolution in permanence”. What Popper insisted on in political systems, Toyota brought to manufacturing systems.
Popper would also have agreed with your post.
Well yes and no DW. Popper is interested in question and answer and iteration. But so were lots of other people. Popper – and his followers are very heavily into the idea of improved approximations to the truth emerging from contest in a free market for ideas. TQM is very keen to reduce the incentives for contest to maximise collaboration and co-operation.
Not that there’s anything wrong with contest in the market for ideas – as they say on Seinfeld. It’s right in the right context, but for better or worse (and it depends obviously enough on the context) at least that aspect is quite contrary to the ethos of TQM. TQM methods are an attempt to square the circle – to provide a way of absorbing many perspectives within some unified view that fits in with central planning.
[…] that the previous post requires as a follow up. A couple of posts ago in this series I discussed feedback within the firm and its growing importance – and the fact that eliciting it well seems as much a social as a […]
[…] I tried to outline in this post, business has pursued various strategies to try to energise the process by which a business […]