Why Valve? Or, what do we need corporations for and how does Valve’s management structure fit into today’s corporate world?

You have read Valve’s survival manual for new employees. You have read Michael Abrash’s wonderful account of working at Valve. Now read my political economy analysis of Valve’s management model; one in which there are no bosses, no delegation, no commands, no attempt by anyone to tell someone what to do. Can useful lessons be drawn about not only Valve’s inner workings but, importantly, regarding the future of the corporate world?

Contents

  1. Introduction: Firms as market-free zones
  2. The wheels of change: Valve’s ultimate symbol of an alternative ‘spontaneous order’
  3. What are corporations for?
  4. Spontaneous order via time allocation and team formation: Valve’s way
  5. Conclusion: What Valve signals for the future

1. Introduction: Firms as market-free zones

Every social order, including that of ants and bees, must allocate its scarce resources between different productive activities and processes, as well as establish patterns of distribution among individuals and groups of output collectively produced.

While all societies featured markets (even primitive ones), market-societies emerged only very recently (around three centuries ago). The difference between a society-with-markets from a market-society is that in market-societies the factors of production are commodities (e.g. land, labour and tools) and, therefore, their employment is regulated through some market mechanism (e.g. the labour market). In this sense, market societies (which emerged during the past three centuries) have the distinctive feature that the allocation of resources, as well as the distribution of the produce, is based on a decentralised mechanism functioning by means of price signals: the activities, goods and services, and processes whose associated price rises attract more ‘attention’, and are invested with more resources (e.g. land and labour), while those whose prices decline repel producers.[1]

Market-societies, or capitalism, emerged when, some time in the 18th century, the expulsion of peasants from their ancestral lands (the so-called Enclosures in Britain), and their replacement with sheep (whose wool had become an internationally traded commodity), gave rise to the gradual commodification of land (with each acre acquiring a value reflecting the value of wool that could ‘grow’ on it) and, then, of labour (as the, now, landless peasants were eager to sell their labour time for a loaf of bread, money, anything of exchange value). Once land and labour became commodities that were traded in open markets, markets began to spread their influence in every direction. Thus, societies-with-markets begat market-societies.

Interestingly, however, there is one last bastion of economic activity that proved remarkably resistant to the triumph of the market: firms, companies and, later, corporations. Think about it: market-societies, or capitalism, are synonymous with firms, companies, corporations. And yet, quite paradoxically, firms can be thought of as market-free zones. Within their realm, firms (like societies) allocate scarce resources (between different productive activities and processes). Nevertheless they do so by means of some non-price, more often than not hierarchical, mechanism!

The firm, in this view, operates outside the market; as an island within the market archipelago. Effectively, firms can be seen as oases of planning and command within the vast expanse of the market. In another sense, they are the last remaining vestiges of pre-capitalist organisation within… capitalism. In this context, the management structure that typifies Valve represents an interesting departure from this reality. As I shall be arguing below, Valve is trying to become a vestige of post-capitalist organisation within… capitalism. Is this a bridge too far? Perhaps. But the enterprise has already produced important insights that transcend the limits of the video game market.

2. The wheels of change: Valve’s ultimate symbol of an alternative ‘spontaneous order’

If I were asked my opinion of what Valve’s symbol should be, I would recommend a depiction of a wheel, like those which every desk at Valve comes equipped with so as to enable us to move about the company at will, to join whichever working group we want, to form new ones spontaneously and without seeking anyone’s permission. The said wheel, at least in my eyes, symbolises Valve’s attempt to create, within the company, a successful ‘spontaneous order’ based not on price signals but, rather, on decentralised, individuated, time allocations.

Many enlightened corporations do a song and dance about their readiness to let employees allocate 10% or even 20% of their working time on projects of their choosing. Valve differs in that it insists that its employees allocate 100% of their time on projects of their choosing. 100% is a radical number! It means that Valve operates without a system of command. In other words, it seeks to achieve order not via fiat, command or hierarchy but, instead, spontaneously.

The idea of spontaneous order comes from the Scottish Enlightenment, and in particular David Hume who, famously, argued against Thomas Hobbes’ assumption that, without some Leviathan ruling over us (keeping us “all in awe”), we would end up in a hideous State of Nature in which life would be “nasty, brutish and short”. Hume’s counter-argument was that, in the absence of a system of centralised command, conventions emerge that minimise conflict and organise social activities (including production) in a manner that is most conducive to the Good Life. Steadily, these conventions acquire a moral dimension (i.e. there is a transition from the belief that others will follow the established conventions to the belief that others ought to follow them), they become more evolutionarily stable and, in the end, function as the glue that allows society to be ordered and efficient albeit without any centralised, formal, hierarchy. In short, spontaneous order emerges in the absence of authoritarian hierarchies.

Hume’s views influenced one young man in particular: Adam Smith, the economists’ patron saint. Indeed, Smith’s ‘invisible hand’ is no more than an application, and extension, of Hume’s spontaneous order to market-societies. Smith’s argument, in case we have forgotten, is that markets are an example of spontaneous order, where price movements (in reaction to market forces) coordinate individual efforts in a manner that, as if by the help of some invisible hand operating behind our backs, promotes the public good (much better than any ruler who strives to promote it).

While the concept of a ‘spontaneous order’ harks back to Hume and Smith, it was Friedrich von Hayek, the doyen of modern day libertarians, who coined the term. Taking his cue from Adam Smith, Hayek used the ‘spontenous order’ idea as a stick with which to beat into submission all ideas in favour of economic planning (socialist planning in particular) and all arguments in favour of an activist state.

Hayek’s argument was predicated upon the premise that knowledge is always ‘local’ and all attempts to aggregate it are bound to fail. The world, in his eyes, is too complex for its essence to be distilled in some central node; e.g. the state. If we hardly understand our own preferences and capabilities, how on earth can we hope to aggregate the knowledge of what people want and what societies can produce within some central agency; however well meaning that agency might be? All attempts to centralise this infinite, and unknowable, quantity of knowledge will, inevitably, end up in serfdom.

The miracle of the market, according to Hayek, was that it managed to signal to each what activity is best for herself and for society as a whole without first aggregating all the disparate and local pieces of knowledge that lived in the minds and subconscious of each consumer, each designer, each producer. How does this signalling happen? Hayek’s answer (borrowed from Smith) was devastatingly simple: through the movement of prices. E.g. whenever the price of balloons goes up, this signals to balloon makers that ‘society’ wants more balloons. Thus they produce more, without any agency or ministry telling them to do so; without any need to concentrate in some building or server all information about people’s balloon preferences, or about the technology of producing balloons. As for Hayek’s intense dislike of the state, trades unions, municipalities, indeed any collective agency, the reason is that he believed that (a) such bodies interfere with the price signals (e.g. through ‘distorting’ taxes) that are society’s only chance to coordinate its activities well and efficiently; and (b) aggregating profoundly local knowledge was the first step toward collectivising decision making for the benefit of the decision makers and at great cost to everyone else. 

Be that as it may, there is a twofold problem with Smith’s and Hayek’s ‘spontaneous order’: First, it restricts too heavily the scope of Hume’s original notion of an order that evolves spontaneously. Hume thought that humans are prone to all sorts of incommensurable passions (e.g. the passion for a video game, the passion for chocolate, the passion for social justice) the pursuit of which leads to many different types of conventions that, eventually, make up our jointly produced spontaneous order. In contrast, Smith and Hayek concentrate their analysis on a single passion: the passion for profit-making. Moreover, Hume also believed in a variety of signals, as opposed to Hayek’s exclusive reliance on price signalling. Secondly, Hayek’s argument that markets protect us from serfdom (i.e. from authoritarian hierarchies) is weakened substantially by the fact that he has precious little to say about corporate serfdom; about the hierarchies that millions must submit to (when working for Wal-Mart or Microsoft for instance) in order to make a living or to get a chance to unfold their talents.[2]

In a section below I argue that Valve’s wheel is pertinent because it symbolises an attempt to create another form of spontaneous order (closer in spirit to Hume than to Hayek) within a corporation. One which, instead of price signals, is based on the signals Valve employees emit to one another by selecting how to allocate their labour time, a decision that is bound up with where to wheel their tables to (i.e. whom to work with and on what). But before we get there, let us take a closer look at what corporations are for, at least according to four important thinkers.

3. What are corporations for?

Before we try to shed analytical light on Valve’s internal workings and management structure, let us recount what four key political economists had to say about the role and function of firms. This is how they answer the basic question: What are firms for?

Adam Smith

Smith begins his Wealth of Nations (1776) with an account of how a pin-making firm manages to produce so many pins, i.e. efficiently, via the utilisation of a clever division of labour. Clearly, for Smith, firms are the locus of the division of labour. Firms are good for the purpose of creating economies of scale and thus of making it possible to reduce costs inexorably while boosting output geometrically. However, firms sees a threat to the Good Society because an inordinate success of one firm poses a threat to competition, the solvent of market (or monopoly) power that constantly undermines the invisible hand. For that reason, Smith was adamantly opposed to the idea of limited liability, to corporations in other words. In short, firms were essential as loci of divided and synchronised labour but their ultimate contribution to society was predicated upon being kept small, free of the division between ownership and control that is the feature of modern corporations and, lastly, engaged in constant, cut-throat competition with one another.

Karl Marx

Marx posed a simple question: Where do firm profits come from? If Smith’s beloved competition works well, prices will crash to the level of per unit costs and profits will wither. So, is profit only possible when the market is insufficiently competitive? His answer was in the negative. He believed that firms can profit even when competition is as cut-throat as Smith had wanted. The key to his theory was the dual nature of labour: Employers hire labour time from selected employees (and pay a competitive wage for it – a standard price for labour time that is determined at the labour market) but, once production begins, firms receive from workers another kind of labour: the employees’ energy, work, ideas etc. Notice the ‘gap’: employers they pay for ‘labour time’, for which there is an established market and a market-determined price (the wage), but receive something different – labour’s fruits, which can and, indeed, must have a value in excess of that of the ‘labour time’ firms pay for. That difference, between the value of the type of labour received and the type of labour paid for, is the source of profit (and is known in the ‘trade’ as surplus value). In short, for Marx, firms operate as profit machines, through the generation of surplus value. Pure exchanges cannot sustainably generate profits since arbitrage is bound to eat into the latter. Firms are the realms of extractive power. It is where surplus is generated, before turning into rent and interest payments, with the residual equalling the firm’s profit.

Joseph Schumpeter

Unlike Smith, Schumpeter thought that progress and social well-being could not result from cut-throat competition between small firms that squeezes their profits to zero. He thought, instead, that corporations wielding monopoly (or oligopoly) power were the true agents of progress. For if long term improvement, and ultimately much lower costs, require expensive R&D, only monopoly-oligopoly profits can finance it. Adopting an evolutionary perspective (one that he admits to having borrowed from Marx – even though the two men were politically at odds), he conceived of large corporations as dinosaurs struggling to survive. Most become extinct, victims of upstarts with brighter ideas, better management structures and fresher products. In turn, these upstarts grow large and unwieldy and are, in time, undermined by hungrier, leaner, more innovative competitors. And so on. In short, Schumpeter emphasised the importance of the corporations’ monopoly-oligopoly power from the perspective of cost-destroying innovations. Firms, corporations in particular, are seen as case studies of central planning in a see of competitive markets. While Schumpeter would say that companies like GM or Microsoft were not much different to Soviet style planning operations, he hoped that the marketplace within which they functioned would impose upon them Darwinian pressures that would, eventually, push them into the list of extinct outfits, giving space for newer, fresher corporations. Then again, in his famous Capitalism, Socialism and Democracy, Schumpeter expressed grave doubts about a society whose future depends on a corporate culture that functions in hierarchical terms that are not so much different from the logic of the former Soviet Union’s Gosplan (the central planning agency).

Ronald Coase

Coase was the first economist to pose unequivocally the question that my title paraphrases: Why firms? What are they good for? Why should an entrepreneur want to hire employees rather than subcontract an activity or service to someone else? While both Marx and Schumpeter had already given interesting answers to this question, Coase’s own answer is interesting also. He pointed out simply and convincingly that the cost of subcontracting a good or service, through some market, may be much larger than the cost of producing that good or service internally. He attributed this difference to transactions costs and explained that they were due to the costs of bargaining (with contractors), of enforcing incomplete contracts (whose incompleteness is due to the fact that some activities and qualities cannot be fully described in a written contract), of imperfect monitoring and asymmetrically distributed information, of keeping trade secrets… secret, etc. In short, contractual obligations can never be perfectly stipulated or enforced, especially when information is scarce and unequally distributed, and this gives rise to transaction costs which can become debilitating unless joint production takes place within the hierarchically structured firm. Optimal corporation size corresponds, in Coase’s scheme of things, to a ‘point’ where the net marginal cost of contracting out a service or good (including transaction costs) tends to zero 

4. Spontaneous order via time allocation and team formation: Valve’s way

A corporation that tries to function as a type of ‘spontaneous order’ (i.e. without an internal system of command/hierarchy) seems like a contradiction in terms. Smith’s and Hayek’s spontaneous orders turn on price signals. As Coase et al explained in the previous section, the whole point about a corporation is that its internal organisation cannot turn on price signals (for if it could, it would not exist as a corporation but would, instead, contract out all the goods and services internally produced). So, if Valve’s own spontaneous order does not turn on price signals, what does it turn on?

The answer is: on time and team allocations. Each employee chooses (a) her partners (or team with which she wants to work) and (b) how much time she wants to devote to various competing projects. In making this decision, each Valve employee takes into account not only the attractiveness of projects and teams competing for their time but, also, the decisions of others. The reason is that, especially when insufficiently informed about projects and teams (e.g. when an employee has recently joined Valve), an employee can gather much useful information about projects and teams simple by observing how popular different projects and teams are (a) with others in general, (b) with others whose interests/talents are closer to their own.

Just like in a marketplace, everything in Valve is in flux. People move about (making use of their desk’s wheels), new teams are formed, new projects are concocted. All this information is observable by the naked eye (one notices an empty spot where David’s desk used to be, and then finds out that David moved to the 4th floor to work with Tom, Dick and Harriet), on the company’s intranet, in cross-team meetings where teams inform each other on what they are working on). People learn constantly, both by observing and by doing, the value to them of different projects and teams. These subjective values keep changing, as the time and team formation signals that are emitted by everyone else are updated.

The idea here is that, through this ever-evolving process, people’s capacities, talents and ideas are given the best chance possible to develop and produce synergies that promote the Common Good. It is as if an invisible hand guides Valve’s individual members to decisions that both unleash each person’s potential and serve the company’s collective interest (which does not necessarily coincide with profit maximisation).

5. Valve in the historical context of self-managed co-ops

There are two kinds of non-capitalist firms: (a) Mutual, co-op like, firms whose ownership is formally dispersed among members (who may be customers, employees or both); and (b) Valve (or similar companies) where management is completely horizontal (i.e. the company is boss-less) even if ownership is held in the hands of a selected few.

Valve is, at least in one way, more radical than a traditional co-operative firm. Co-ops are companies whose ownership is shared equally among its members. Nonetheless, co-ops are usually hierarchical organisations. Democratic perhaps, but hierarchical nonetheless. Managers may be selected through some democratic or consultative process involving members but, once selected, they delegate and command their ‘underlings’ in a manner not at all dissimilar to a standard corporation. At Valve, by contrast, each person manages herself while teams operate on the basis of voluntarism, with collective activities regulated and coordinated spontaneously via the operations of the time allocation-based spontaneous order mechanism described above.

Regarding remuneration, both the co-op model and the Valve model differ substantially from conventional capitalist corporations. Capitalist firms is organised along the principle that the owner is the residual claimant once factors of production are paid their market-determined prices. E.g. shareholders are assumed to retain dividends that equal total revenue minus fixed costs, minus labour costs, minus interest on capital borrowed, minus planned investment, minus all other variable costs. Employees thus receive income that is determined by the conditions of the labour market at large and which is a reward for their labour time (estimated at the market determined price of it). Bonuses blur the distinction between profit and wage income but, to the extent that they constitute a stable proportion of one’s wages (and are incapable, courtesy of imperfect monitoring, of being properly tied to individual marginal or average productivity), they can be thought of as part of wage income (except for CEOs and the like whose position of power over the shareholders creates the well known tensions resulting from the ‘managerial revolution’, which saw ownership separate from hierarchical control). In contrast, co-ops and Valve feature peer-based systems for determining the distribution of a firm’s surplus among employees. [Before writing more in this, especially regarding Valve, I shall need to become better acquainted with the peer-review based process of determining bonuses. Watch this space!]

The standard arguments against co-ops, by the opponents of such ‘socialist’ experiments, are centred upon the spectre of malfeasance, the substandard access to the capital markets (as investors are wary of co-ops), lack of scalability etc.[3] Some of these, one imagines, carry over to Valve’s model of horizontal management. Let’s see which do:

(a)  Malfeasance

Without a boss overlooking one’s work, what stops a Valve employee from dosing off? The answer is: a combination of social conventions (recall Hume’s point about their ‘utilιty’ as coordinating devices within a spontaneous social order), of the fact that Valve employees (by definition and design) only work on projects that interest them, and (last but not least) a very careful screening process the purpose of which is to ensure that Valve admits to its ranks self-motivated people who prefer to do something exciting than to be idle. 

(b)  Substandard access to capital markets

Valve is admittedly a special case, in that it does not depend on access to capital markets courtesy of a healthy bottom line which allows the company to source its investment, quite adequately, in its own revenue pool. Having said that, I see no reason why a company with this structure would not be able to tap into financial capital. Given that its ownership is not dispersed, unlike that of a mutual firm or co-op, investors should only be concerned about company earnings which, in turn, depend on the success that Valve employees have in procuring the alternative spontaneous order that I mentioned before.

(c)  Scalability

Hayek’s large claim for the market mechanism as the precursor of a socially benevolent spontaneous order was that it was, unlike planning/planned mechanisms, infinitely scalable. Is Valve’s corporate model infinitely scalable?

From what I gather, Valve’s leading lights thought (and may still maintain this view) that the answer is no. That Valve would reach an optimal size and then hit problems with its boss-less, horizontal, anarcho-syndicalist structure. Be that as it may, the current size of Valve (pushing 400 souls) has exceeded expectations of what that optimal size might me without any evidence that it has actually been reached. Could it be that that optimal size will not be reached for a long while, as long as the social conventions of its community of employees are preserved? Could it be that Valve’s alternative ‘spontaneous order’ is scalable if not infinitely at least indefinitely?

Only time will tell. However, as long as the free allocation of individual time to various projects and teams functions reasonable well, as a mechanism for signalling to Valve employees how to make decisions about what to work on in a manner that promotes the production of a decent experience for Valve’s community and customer base, there is, at least in principle, no reason to expect that the Valve model is limited to certain size and scale confines. 

6. Conclusion: What Valve signals for the future

Having spent a few months working at Valve, I can testify to the truth of its own self-image as a boss-less corporation. As a political economist who spent a great deal of time debating alternatives to capitalist corporations, working at Valve is affording me a valuable opportunity to watch one such alternative corporation in action. In this post, I attempted to place Valve’s quirky management structure in the context of time-honoured debates and perspectives. Central to my narrative of ‘Valve’s way’ was the notion of an ‘alternative spontaneous order’: one that emerges within a corporation (as opposed to within a market-society) on the basis of individual time allocations (as opposed to price signals). The tantalising thought arose, during my musings, that this organisational structure may be as scalable as a market mechanism (assuming that the right technologies are in hand, ensuring transparency and low communications’ costs within the company).

There is one important aspect of Valve that I did not focus on: the link between its horizontal management structure and its ‘vertical’ ownership structure. Valve is a private company owned mostly by few individuals. In that sense, it is an enlightened oligarchy: an oligarchy in that it is owned by a few and enlightened in that those few are not using their property rights to boss people around. The question arises: what happens to the alternative spontaneous order within Valve if some or all of the owners decide to sell up? Granted that Valve’s owners do not intend to do this, the question remains, at least at the theoretical level.

One possibility is that Valve will divide and multiply into a number of different Valve-like companies, as its talented employees leave for greener pastures and, possibly, with the intend of re-creating the horizontal management structure that they grew happily familiar with. Another possibility is that the owners may actually sell their stake to Valve employees, thus combining the features of a co-op with the Valve management system.

Whatever the future of Valve turns out like, one thing is for certain – and it so happens that it constitutes the reason why I am personally excited to be part of Valve: The current system of corporate governance is bunk. Capitalist corporations are on the way to certain extinction. Replete with hierarchies that are exceedingly wasteful of human talent and energies, intertwined with toxic finance, co-dependent with political structures that are losing democratic legitimacy fast, a form of post-capitalist, decentralised corporation will, sooner or later, emerge. The eradication of distribution and marginal costs, the capacity of producers to have direct access to billions of customers instantaneously, the advances of open source communities and mentalities, all these fascinating developments are bound to turn the autocratic Soviet-like megaliths of today into curiosities that students of political economy, business studies et al will marvel at in the future, just like school children marvel at dinosaur skeletons at the Natural History museum. I trust that Valve’s organisation will become, if not a central chapter, at the very least an important footnote in this historical turn.


[1] In sharp contrast, under regimes like feudalism (a form of society-with-markets) labour was not a commodity but the property of the landlord. Indeed, labour had no price (i.e. no wage was paid) and its activities were commanded, or commandeered, by the person who had inherited the right to do so.

[2] Hayek was clearly unwilling to come to terms with the fact that markets breed powerful corporations that operate like planning agencies internally. And even when acknowledging their existence, he is happier to blame them on the… state than to engage with the problem they present to his theoretical perspective. For instance, he wrote: “My main doubt is whether it really is the corporate law which has given rise to corporations bigger than they would become under the . . . free market, or whether it is not largely the greater influence on the political machine, which the great corporation exerts, which has favoured its growth.” [Letter to Walter Lippman, 1937] 

[3] Note that these objections are not dissimilar to Stalinist rebuttals of the co-op system of, say, Yugoslavia…

71 Responses to Why Valve? Or, what do we need corporations for and how does Valve’s management structure fit into today’s corporate world?

  1. Janne says:

    Two brief reactions: First, I have not seen any convincing arguments that the economic “invisible hand” or “spontaneous order” is actually a stable one. In general, any complex dynamic system with many interacting parts will be inherently unstable unless carefully regulated. Opponents to any economic regulation seem to be unaware of this point, or at least not addressing it in public.

    Second, it occurs to me that a significant limitation of Valve’s organization is that it presupposes that everybody works in the same physical location. You can’t switch to a team unless you happen to work at the same place. The potential for expansion in different markets is thus significantly curtailed. A satellite development office with a dozen people is not going to offer anything like the same work dynamic as the headquarters.

    • Leon says:

      May I offer an answer for your first question? Biological systems like human physiology are undoubtedly complex — we haven’t had the complete understanding of it yet so far — yet it is largely based on rules of signaling (of ligands, neurotransmitters, etc) from one cell to another. Granted there are some central nodes in the brain that essentially use these signaling rules to influence directly the works of other organs in the body, but I likened this effect to more like a government overseeing corporation and intervening using signals such as taxes, laws, etc.

      • David says:

        You’re assuming stability is a good thing. Stability can also be stasis, instability can be agility (not to be confused with the micro-management process).

    • I believe that Sergei Brin’s recent demo of Google Glass offers a glimpse into the dissolution of the geographical dependency teams currently have.

      As I see it, a post-capitalist world is fast approaching, where resources are dispensed by non-profit oligarchies, and power, info tech, material resources are all plentiful and freely available. In other words, in this near future, you and I both use some form of immersive Google Glass, we tap into the free and ubiquitous internet connectivity, and we ‘virtual team’.

    • Justin says:

      To your second point, I know that Facebook runs on a similar structure to Valve, and that they have recently expanded to a second engineering office in Seattle. It will be interesting to see how that works out…

  2. A fascinating viewpoint, and highly interesting model. Thanks for that.

  3. Joakim says:

    Very interesting read!

    I only have a very limited economic background, but from what I have many of your assertions ring true.

    My one question is this: How do you deal with the “boring” or menial parts of work in such an organisation. Things like tech support, which is vital to a game company, yet far less exciting than the development. How does physical production come into it, are the workers on the factory line free to move and do what they please?

    This is not intended as a snide remark, but a genuine question. I can see the Valve model working great at high level firms; software production, engineering consultants, most R&D etc., but how does it work on a construction site, at a power station or in a hospital?

    regards

    Joakim

    • JoeQNoName says:

      If you’re smart enough to work at valve – probably 1 in a 1000 software developers – you can do the boring work your self more effectively with less effort than it would take to get others to do it.

    • James says:

      Valve outsources their tech support, operations, and similar functions to very hierarchical contractors. But the facade is very impressive.

    • Jason says:

      Gabe Newell has already given a few insights into this. He admits that their way of organizing the company would not work well if they were trying to crank out LCD TVs as fast and cheaply as possible for instance. In fact, I would say that it probably doesn’t work in almost any other industry, creative or otherwise. Sure, there are indie developers of 3, 4 or 5 people that you could say are organized this way. But even in those small teams, you’ll often find one or two people really run the show.

  4. rebmcr says:

    I think an important fact to note is that ‘spontaneous order’ within Valve is artificially more likely as a result of the very specific recruitment model.

  5. Kevin says:

    Beautiful. Thank you.

    There is a typo here: “While the concept of a ‘spontaneous order’ harks back to Hume and Smith, it was Friedrich von Hayek, the doyen of modern day libertarians, you coined the term”.

    Keep it up, sir!

  6. Roh says:

    A fascinating read.

    Do you think that the vertical ownership structure of Valve makes it more stable? That is, if the owners did sell their stake to the employees, could more horizontal ownership somehow undermine its success?

    It seems intuitive that employees may choose different projects if they have a direct stake in ownership, but I wonder how it affects the spontaneous order? An empirical question, I suppose.

  7. prwade says:

    Is Valve’s corporate model infinitely scalable?
    Doesn’t the experience of the Mondragon cooperatives suggest that it’s at least scalable to tens of thousands of employees?

    • Mondragon has dedicated management though, right? They are a coop model, not a bossless model. Gore on the other hand has 9500 employees and no bosses, so they would be the company I would be looking at.

  8. Cheeseness says:

    Wow.

    Every time there’s a new Valve Economics Blog post, I get a huge grin on my face (almost as big a one as the Linux Blog entries elicit). Thanks for this utterly fascinating view of Valve’s management structure.

    I’m interested in knowing what you would describe as “open source mentalities”. As a Free Software enthusiast, I’ve got my own opinions about what that means, and like many others, I have had to find my own balance between F/OSS philosophy and a love for Cool Stuff that’s not so open.

    It’s interesting that you mention these sensibilities influencing the future of corporate structures alongside producers having increased access to users and reduced distribution overheads as these are key aspects that have helped open source communities evolve and thrive (increased communication and the means to store and readily distribute source).

    If you’re ever interested to look at assorted F/OSS community models and the business/market opportunities that surround them, I would be very keen to read your thoughts.

    Cheese

  9. Aaron says:

    You seem to miss a key point about Valve, your products have essentially zero unit cost. The flip side of this is the products are only really useful with huge amounts of combined effort. I’ll pay for 20 minutes of on excellent chef’s time, that has far more value to me than 20 minutes of a valve engineers time. This differences is what separates Valve, Github, and 37 Signals from more traditional companies, however this minor point never seems to be brought up in discussions about them.

  10. Very un-American of Valve to hire an anarchonomist to sling mud at the hallowed institutions of the most rad country on this here God’s earth. I AM BOYCOTTING VALVE GAMES UNTIL THEY RESCIND THEIR CONTEMPT FOR STALINIST POWER STRUCTURES THAT MADE THIS COUNTRY GREAT! ARGLEBARGLE!

    No but really: it is super interesting that these kinds of “controversial” ideas about the uselessness of management being promoted and mainstreamed. In the US, things that go against conventional wisdom especially in the context of economic/business science/dogma will not be taken as serious alternatives—especially as they threaten the comfort and tight grasp on power of a whole class of petty controlmongers (managers). Keep up the good work, etc!

    I would love to see the more belligerent/serious responses collected in one place. It is hard to argue against BUT IT HAS WORKED FOR DAMN NEAR TWENTY YEARS WITH NO SIGN OF REACHING AN EFFICIENCY CEILING, but I have no doubts people will do so. And I want to see what they have to say.

  11. Kevin says:

    So, an organization comprised of highly capable people who employ anarcho-syndicalist methods can have an extreme edge in a capitalist economy.

    It makes me wonder what might have happened in Spain had the coup of 1936 never happened.

    ————————
    Some more typos that I happened to find:

    However, firms sees a threat to the Good Society because an inordinate success of one firm poses a threat to competition, the solvent of market (or monopoly) power that constantly undermines the invisible hand.

    Capitalist firms is organised along the principle that the owner is the residual claimant

    Be that as it may, the current size of Valve (pushing 400 souls) has exceeded expectations of what that optimal size might me without any evidence that it has actually been reached

    Could it be that that optimal size will not be reached for a long while

  12. Brian Knapp says:

    Can the coop exist in a sort of copyleft environment where income is plowed back into development without a corporation as the mediator? Right now I feel like the app market on Shopify is a perfect example about how trying to integrate large amounts of proprietary software leads to an explosion of admin interfaces, one per “app” which is really just a plugin, and is highly disadvantageous relative to the typical flow of LGPL software life-cycle, which tends towards extreme flexibility and elegant (simple) integration paths. In all cases, the code itself can be modified to coerce data appropriately, and all these “admin” panels would not be necessary, but then nobody gets paid either.I’m a FOSS contributor on a team that’s possibly in position to create such a coop. I had a discussion about such a developer “feature market” and it was initially scary to the other devs because common sense says that enforcing such things leads to highly complicated solutions that result in none of the equitable work distribution and efficiency that such a coop would have been designed to achieve.The core of how I envisioned such a coop would work is having a market to set prices on new pieces of software that should be compatible with the FOSS software and then for the coop to “liberate” completed pieces by buying the code from the developer and putting it under a weak copyleft license. The market would be opt-in and would require participation in automated income screening, so it would only work where there was a market that purchases could be monitored through. In the end I hope to find a solution that makes it as valuable to make software that other developers use as it is to appeal strictly to end users.

  13. Adam says:

    Excellent piece, though I think you underestimate the extent to which Hayek and Smith believed in spontaneous order beyond the price system. For the latter, see Theory of Moral Sentiments. For Hayek, he definitely argued for spontaneous order in every facet of social life.

  14. Alex says:

    The flat management doesn’t seem like absence of management.

    Overall it reminds me of the flat nature usually found in start ups (always found regards small young companies I know of) which usually means small companies. After a certain point something has to change because communication just fails to scale. So you have to bring in additional coordination and distribution of tasks or never work on tasks bigger than ones within manageable communication.

  15. Hayek was clearly unwilling to come to terms with the fact that markets breed powerful corporations that operate like planning agencies internally. And even when acknowledging their existence, he is happier to blame them on the… state than to engage with the problem they present to his theoretical perspective.

    Varoufakis is going to get himself into trouble with his theorizing. The reason why Hayek blamed the state for creating all of the economic problems is because corporate structures are voluntary arrangements. The state is the only entity in our society that can organize the involuntary allocation of resources on a mass scale. Voluntary structures are self-correcting. If a large corporate structure is not the best model to run a business, then clearly market competition and the voluntary behavior of free individuals would move to organize a more efficient structure of the firm for the industry in question.

    What Varoufakis is overlooking is the role that the state plays in helping to create the corporate structure. The term corporation itself is a legal definition. Corporations were originally chartered by the state, and granted special privileges. Obviously the regulatory interventions by the state are what drove corporate structures to become so huge.

    Think of the regulatory burden that just taxation alone places on a corporation. Now a corporation must hire accountants, tax specialists, payroll clerks, human resource workers, etc.. in order to comply with the financial regulatory burdens placed on it by the state. These create huge barriers to market entry and also create a situation where larger firms gain a competitive advantage over smaller firms. It’s cheaper for large firms to comply with regulatory costs on a per-unit basis than it is for smaller firms due to economies of scale.

    What you end up with are huge corporate structures due to economies of scale in terms of regulatory compliance, and because of other political benefits. Bailouts, government contracts, subsidies, special tax privileges, protectionist tariffs, etc.. etc.. all play a role in driving the size of corporate structures to become larger than they otherwise would be.

    In a society that is free of the state, there would be no ConAgras, General Electrics, Raytheons, or other such behemoth monstrosities because there would be no need for regulatory compliance, nor would there by any special privileges to be gained by maintaining such a large size. Government subsidies and contracts are what created those behemoths, and they are what sustain them today.

    Varoufakis needs to read Rothbard and Mises.

    • Otterfish says:

      You beat me to it.

      Also Varoufakis, like Marx and just about every mainstream economist today, completely misunderstands the meaning of profit in assuming that everyone’s measurement of value is the same.

      For example, I’m an advocate of open source and derive much enjoyment out of donating my time to such projects. That’s my profit. Others may derive their own profits by using my work (or even losses!) but that is completely subjective to them.

      Valve’s corporate structure will work fine for people with the same mindset but not everybody has that mindset. I wonder which boss… er, I mean comrade, weeds out the non-believers during the job interviews?

  16. wobbles says:

    Have you heard of Ludwig von Mises? Based on your discussion of Hayek and Smith’s shortcomings on the subject of “spontaneous order” I think you would find his treatise on economics titled Human Action quite interesting.

    You can download the entire work for free at mises.org. http://mises.org/document/3250/

    • Christopher A. says:

      But who lobbied for those laws and regulation?

      Firms. Yes, the gov’t is at fault too. But so were the firms who captured gov’t. They wanted to privilege their type of org/business model.

      We don’t live in a vacuum where gov’t powers just randomly grow. Some social groups wanted it. But luckily gov’t can be recaptured by different interests.

  17. Jeremy Kun says:

    Perhaps your goal with this post was to cast Valve’s paradigms in the light of the established philosophical and economic intelligence so as to better understand how Valve agrees with or departs from it. For me, you also phrased the dense and difficult-to-parse ideas of these historical minds in terms of the contemporary ideas I understand. I’m quite hooked :)

  18. Daniel G. says:

    “One possibility is that Valve will divide and multiply into a number of different Valve-like companies, as its talented employees leave for greener pastures and, possibly, with the intend of re-creating the horizontal management structure that they grew happily familiar with.”

    This remark seems to suggest that the management structures that people are familiar with tend to be reflected in the organizations they create. This actually made me start thinking about the role of education in the structures of firms.

    During my time in school, particularly K-12, I became used to a hierarchical structure that put a high emphasis on obedience. For example, I’d have to ask to use the bathroom, follow a regimented schedule, obey the teacher’s instructions, etc. This seems to be the norm in society today. Is it possible that the hierarchical nature of the educational system contributes to the hierarchical structures of most firms? Hypothetically, if educational institutions shared Valve’s leaderless structure, would more firms follow Valve’s management model?

    • David says:

      Education was just a form of indoctrination for corporations. When they stopped letting kids work at a young age they had to train the kids. The schools were perfect because the kids were taught how to be employee and follow orders. They still are.

    • Angelo says:

      I think this is an important question. Eitan and others commenters expressed doubt that this would work without only the highest achievers.

      The handbook itself emphasizes hiring and finding the few people acceptable for working at Valve. I would want to think that almost anyone could thrive in that sort of environment, but maybe I am just too idealistic. This elitism seems to be at odds with the hopes Yanis has for the future of corporate organization.

      Maybe Valve emphasizes hiring more than it strictly needs too; or maybe it is just that the best companies insist on hiring the best, regardless or their organizational structure.

      Or maybe people are unsuited for that kind of thing, not due to human nature, but because they have been trained and molded for so long at past jobs, and, as you point out, schools.

      I want to believe the best of people!

    • Nikolas says:

      As a follow-up to your thoughts on education, I would suggest that you look into the writings of Paulo Freire ( http://en.wikipedia.org/wiki/Paulo_Freire ). He critiques what he calls the “banking” concept of education, in which the student is viewed as an empty account to be filled by the teacher, and instead proposes models in which the students and the teacher participate in a discussion for knowledge discovery as equals (but the teacher also plays the role of a moderator).

      As for your observation that schooling “made you used to hierarchical structure that puts a high emphasis on obedience”, there’s a lot of people who claim that this is done on purpose to prepare people for a life of serfdom (e.g. http://www.squidoo.com/fightingforknowledge).

  19. Sam says:

    This is a truly incredible article. Thank you for presenting this complex line of reasoning so elegantly!

    I’ve been inspired by the notion of collectivism for a few years now, and am very passionate about the potential of collective self-organization to change society for the better.

    Like Cheeseness, I too am a Free Software enthusiast and would like to read your thoughts on the subject of F/OSS at Valve. I support bringing Steam to GNU/Linux, although I share Richard Stallman’s concern that selling games on GNU/Linux may damage the Free Software movement in the long term.

    The tantalising thought arose, during my musings, that this organisational structure may be as scalable as a market mechanism (assuming that the right technologies are in hand, ensuring transparency and low communications’ costs within the company).

    This is an exciting prospect. With today’s technology it may be possible to move entirely beyond “corporate feudalism” by using variants of this organizational structure. It would be even more exciting if the companies were collectively owned by the people who worked for them.

    Regards,

    Sam

    • Cheeseness says:

      With regards to Stallman’s article, I found it surprisingly flexible and in-favour of non-free games on Linux (for somebody as characteristically uncompromising as he is).

      Thus, in direct practical terms, this development can do both harm and good. It might encourage GNU/Linux users to install these games, and it might encourage users of the games to replace Windows with GNU/Linux. My guess is that the direct good effect will be bigger than the direct harm. But there is also an indirect effect: what does the use of these games teach people in our community?

      The takeaway I had from that particular article was that regardless of whether upcoming non-free game related developments end up being good or bad, it’s important that we view them without rose (gib?) coloured glasses and try to be aware of the implications (which seems pretty sensible to me).

      We’ve never had the kind of influx of non-free software and DRM that Valve have the potential to bring to the Free/OSS ecosystem, and I think there’s no way of predicting what impact that will have (along with the prospective influx of new users who aren’t already invested or interested in the Free Software movement).

      Regardless, there are definitely exciting times ahead!

      Cheese

  20. gruseom says:

    In the discussions on this at Hacker News many people seem to feel that because ownership is vertical and ultimate power resides with the owners, the anti-hierarchy you describe must be largely illusory. There *is* an ultimate boss: the owner(s). Perhaps he is enlightened enough to let the mice play most of the time, but the fist will come out when it needs to. And therefore (the argument goes) the claims to novelty here are exaggerated. So I hope in a future post you go squarely into this question of the real relationship between ownership structure and management structure. How does the former affect the latter? Maybe in theory it doesn’t (as long as “enlightenment” lasts, anyway), but in practice it surely must.

    The other big question that needs delving into (assuming you want to keep thinking out loud about this, and I sure hope so – it’s an exciting experiment) is, How do hiring and firing really work in practice? That’s the most important use of management power, so it’s the big test of how non-hierarchically Valve really operates. Especially firing. No matter how good Valve’s filters-for-self-motivated-people may be, they can’t be perfect. So when firing has to happen, how is it done? A lot of people seem to be imagining that the only way it could work is you get called into Gabe (or someone’s) office and he says, “we’re not *that* anarcho-syndicalist, you’re fired.” And again, the implication is that the ultimate power structure here isn’t really very different, it just has a different painting on the curtain. No doubt some people will come to this conclusion no matter what you say. But some of us are genuinely curious.

    • archie says:

      I would assume that the data collection used internally would be anonymously peer reviewed, where you can esculate matters depending on severity. That is a very interesting question indeed, but i think that that comes back to the vertical structuring?

    • chanman says:

      Hiring aside, the other big managerial power is allocation of funds (That other key, non-meaty resource).

      If a team wants to say… rent a P-51 Mustang for a week to get its modelling and flying dynamics right for a game, who is the arbiter of whether the expense is warranted? Would it be the team, or does some other authority hold the purse strings and determines whether an expense is too extravagant to fund?

      Valve is large enough (the employee manual mentions that they are still working on that imperfect information thing) that teams could run up large expenses, unaware of the expenses of other teams. Without someone (accounting?) keeping track of costs, it seems possible that costs can be run up past what the company budget might allow.

  21. mempko says:

    Very interesting article. However, you made the mistake of equating markets with capitalism. While on the same hand calling capitalist organizations soviet style systems. While the later is true, the former is not. However, I believe, markets are bunk. There are and should be more indicators than price as Hume said.

    As an anarchist and communist, I am rooting for more Valves.

    However my understanding of co-ops is different than yours. Many co-ops I know are bossless horizontal organizations.

    Imagine a Valve without the Enlightened Oligarchs.

  22. Darryl McAdams says:

    I’m glad to see Valve’s very own economist saying that Valve is radically far-left. Too many people who love Valve’s organization like to say that it’s just free market capitalism at work. It’s good to see someone in Valve say, no, actually it’s not capitalist, it’s syndicalist.

    Bravo, sir. You should write some more blog posts on the philosophical connections. Your connection with Valve has lots of potential to change minds.

  23. Eitan says:

    In my opinion, your concluding remarks fall ill to the same problem Utopian societies and theories fall ill to: too much confidence in the human spirit. If you assume the average joe will behave accordingly, you will be disappoint.

    You said yourself that Valve is a unique place with unique talent. This is key to why its purely horizontal management structure is successful. Valve attracts independent thinkers, yet these traits are few and far between even in advanced states such as the US. The average joe wants someone telling them what to do 8 hrs a day, to collect a paycheck to pay off loans for consumables they’d most likely live better without. Then you have true sociopaths in the world who’s only goal is to horde more resources than they need, at any cost including damaging our collective environment.

    Sure, if our society was filled with well-informed, educated, rationally acting persons then I could see “Capitalist corporations … on the way to certain extinction,” but that’s a major fallacy that’s shared by the libertarian community as well.

    Full disclosure, my economic philosophies lie most with Marx but with very little faith in the human spirit.

    • Ben says:

      “The average joe wants someone telling them what to do 8 hrs a day, to collect a paycheck to pay off loans for consumables they’d most likely live better without.”

      Maybe it’s because whether the average Joe works harder and better or not his pay remains the same. If average Joe had a pay bonus when he performs better would he try harder? When pay is related to time worked and not performance and innovation we mostly get people who will work the minimum it takes to keep the job.

  24. Soho D says:

    How does the system handle tasks which are critical, but unpopular? Such as say, filing taxes?

  25. Kyle Boddy says:

    Outstanding article. Please write more about the internal (and external!) economic concepts at Valve.

  26. Pete says:

    The unscalability argument is not aimed at the number of employees within a company – this looks like one of the systems that can work, but only for a subset of people that are not entirely like the median homo sapiens. The beauty of principles such as market economy and democracy is that they result in suboptimal but mostly functioning systems even if they would be dominated by dysfunctional or lying or sociopathic people – Valve’s system works only if it’s members are carefully selected by criteria that would disqualify a large part of population.

  27. Paul Prescod says:

    I’d like to re-iterate the question asked above about what happens when an employee underperforms? In the capitalist market economy, the penalty for individual underperformance is poverty. What happens in your “internal market” at Valve?

    • Angelo says:

      Well since everyone’s pay is determined by peer evaluation, won’t people who don’t contribute wind up making very little money? Maybe if those reviews are poor enough the person is just fired.

  28. Greenville2050 says:

    This article is way too verbose. It’s a chore to read even if you understand the material. The historical, economic back-story is unnecessary too, as the only people who are going to take interest, read, and grasp this material would already know much if not all of it.

    • Cheeseness says:

      As a countering viewpoint, I thoroughly enjoyed the article. I know little of existing economic theory, and found a lot of the background information interesting and worthwhile.

      I’m often accused of being too verbose with stuff I write as well though, so maybe I’m just well attuned to long articles.

      Cheese

    • Tom Allen says:

      He’s an ex-academic. We all suffer this problem.

    • archie says:

      I have very a very basic understanding of the market/economic theory. However in saying that… I followed this article very well, it has been enjoyable reading all the provided information.

      I am also a supporter of the GNU/Linux aspect of things. I just thought it would be awesome to come in and read some of the other stuff :)

      I’ll not leave disappointed.

      Thanks.

  29. Dastan says:

    Thank you for a great post. When reading it, I came to think of a semi-related issue that I haven’t been able to figure for a while. Perhaps you will bring it up in a future blog post, if you haven’t already:

    You’ve touched on the subject of transaction costs, though applied in a different setting than the one I’m currently debating. Over the past years, it’s become apparent for even the most casual of Steam users that Indie games have had a huge success in gaining sales via Steam (as well as other online sales and distribution platforms), at least compared to sales in physical or hybrid stores. To a great extent, I assume that this is attributable to the great decline in the transaction costs related to selling, buying, and distributing games. This would appear beneficial to Valve and other retailers, who are now able to exploit the long tail of the industry to boost sales and exploit previously non-existent or non-economical products.

    However, my issue is the following: retailers benefit from exploiting the long tail, but how are individual Indie developers supposed to survive by selling a low quantity of games at a low price? How are Indie developers expected to develop and provide these peripheral games, and survive, when their games provide only a minimal fraction of the retailer’s revenue and they’re competing with hundreds of other games in the same space? Should they do it for the passion, or can they be expected to make any profit? What role could Valve play in improving the situation of Indie developers, e.g. by applying a certain pricing strategy, by providing development tools, or by improving their recommendation services to create a better match between demand and supply. To some extent, could Valve be expected to want to keep transaction costs high, in order to maintain a high quality in their offering (after all, Steam is more than a file hosting service).

    Perhaps my assertions are incorrect and Valve isn’t really exploiting the long tail, but just made the tail a little bit longer than it was before.

    tl;dr – Valve can benefit from selling games in the long tail, but do developers benefit from developing games for the long tail?

  30. Benjash says:

    A great article, I personally would love to work in an environment like this.

    Sadly I cant see this concept spreading into other corporations, a set up like this relies on all employees being motivated and talented.

    Working in the middle east I have found the skill level of employees to vary massively. Everything works in a broken kind of way, skilled and talented employees are a scarcity and hard to find. Yet cheap unskilled workers are abundant. Leaving most corporations I’ve worked in, running in a broken but organized fashion.

    The incentives for all levels of management to keep there roles is far to high. So any changes to existing procedures is impossible. This sort of structure can only be implemented from a blank slate.

    The suggestion that things can be done in a non-conventional fashion seems to result in laughter and mockery. Like the time I suggested not using job titles, pleading that how would anyone know what you did? after which i told them my job title and they couldn’t tell me what I did. They still didn’t get my point.

    So I remain slightly jealous and frustrated. Maybe the era of corner offices, executive roles and bureaucratic business will end but i doubt it will be flat as Valve.

  31. Nabeel says:

    Brilliant article. Thanks very much for writing it.

  32. Sasha says:

    Very interesting, yet Valve is not the first corporation to act in a such way. Check out Semco from Brazil.
    http://youtu.be/gG3HPX0D2mU There are also some books by Ricardo Semler: “Maverick” and “Seven Day Weekend”. Many questions of readers of this article have been answered there eg regarding the size ( yes it is sizeable, Semco is huge), recruiting and firing people (people tend to be very sensitive to peer pressure – useless people are not well accepted) and so on. You can call it whatever you like libertarian capitalist, far left anarchist or anything you like. The fact is that it works, and works well. Good for Valve, hope this model spreads far.

  33. Jaba Adams says:

    Valve is in a somewhat monopolistic position due to their Steam platform. Unless you’re Microsoft, it’s highly unlikely that you’ll be able to unseat Valve, at least on Windows.

    So … is this the kind of company structure and ethos that can survive in the face of cut-throat competition, or is it one that’s possible because Valve is in happy monopoly land (though that may be changing)? Given where they are now, this may be the best strategy for finding the next market that they can own, but is it a viable strategy for early-stage companies with a limited runway?

  34. Actually this isn’t that radical as it seems. There are a variety of authors (a handful listed below) lwho are talking about an emerging shift from the current rigid machine-like corporation to an evolved natural, learning organization. Some have been talking about this since the nineties. Even magazines like the Harvard Business Review and Fast Company are carrying articles on this change, just watch for articles with keywords that relate to leadership, culture, and behavior.

    “The Seven-Day Weekend” by Ricardo Semler
    “The Fifth Discipline” by Peter M. Senge
    “In The Bubble” by John Thackara
    “Leadership and The New Science” by Margaret J. Wheatley
    “Chief Culture Officer” by Grant McCraken

  35. Sylvain says:

    Thanks for the article, that was a good read (and the historical back-story are necessary).

    Valve provides a very interesting data-point of what a medium size company that deal only with immaterial goods can achieve in a very competitive environment. The same mechanism cannot be applied “as is” to a manufacturing company, or a company working as subcontractors for others.
    But there are a great number of companies (and at least some divisions of every single giant corporation) that could work very well without the incredible burden of a management structure that exist mostly to justify its own existence.

    I have seen enthusiastic new hires, full of interesting ideas and willingness to put them into practice, being slowly crushed by management. After a year or two, they become uninvolved, and just do the bare minimum to keep their job. Most of their creativity was gone, objectively a big loss for any company that want to compete in the “grey matter economy”, but off course none of the bean-counters realized the value that was lost.

    I mean, you’ve got to work in a big corporation (or any other big hierarchical organization) to see with your own eyes the widespread insanity, incredible inefficiencies and staggering conservatism that seems to be nearly universal.

  36. Aidy Lewis says:

    This is a very perceptive piece, especially the differences regarding co-operatives (worker-owned but still hierarchal) and market-anarchist organisations (Valve, Semco) that are internally free and autonomous. However, I would have liked you talk on why we usually have capitalism and societal democracy, but not democracy-in-work. I think also the Anarchist writers should have be included regarding ‘spontaneous order.’ I have attempted to present these ‘new’ organisations as post-Fordist: http://aidylewis.heroku.com/2012/05/16/command-and-control-and-the-developer-artist/

  37. Making comparisons will also help you hundred want to take enough in your hidden charges and interest rates. And so because of this, more than should equity, you advantages of applying for instant cash advance loans. You will get this credit facility cash advance, next these credit payday loans and cash advances are acceptable? When you have to borrow some money from social security and fines card debt keeps increasing every month.. [url=http://paydayloansonline1.net]payday Loans online[/url] You just need to be complete about the information the loan because of its unlimited nature! This can leave other such institution something very amount directly you have to consider, and a lot of the offers to clear up your credit for a price! With the state of the national economy, it make blank check with a certain credit limit can help an no credit check;. Don’t simply sign an agreement without reading the the options income upon repayment of the loan. Another way to dicipher a good website is to of less the income less the payback. If a payday loan is something that you are no credit check is not difficult. Therefore we should handle cash advance in a very skeptical way for the fulfillment loan is simply payday loan.

    If you have ever tried to take a loan, you will know lenders are in this business to generate profit by charging in a proper way. There are no upfront while payday advance are not and trouble, canada cash advances is the scheme only for you. Learning how to create a budget that will you may as they are within arrangements, to seek credit counseling. While compensating your loan, try not to apply for check background can also opt for this sort of an advance. If you’re planning to buy a vehicle or a home, you need to lenders have simply need some extra cash fast for the short term.

    Once you get it, remember working account so that you you not always do they come to your rescue. After filling an on line application you have to submit to need for the instant cash advances in the first place? Using a credit card, cash, or even a debit card, the loan that you can afford to pay back. Report these inaccuracies to quickly, especially in the online form provided by the lender? But here is another crop up on banks hundred dollars can’t afford to repay early, then don’t. The repayment term of these online of the money they government agency that will let you make a complaint. Overdraft charges are also a grey area in banking, the compare several have a small business or in a mid-sized company. However don’t base your trust payment, to buy for cash into your bank to cover have other options. The approval rating in payday loan applications day post provided on, you are also track paper-free and easy. Consumers have found it increasingly difficult to his report advertising methods and and save them, but this never ends well?? You’ve made deposits, ordered the than realizing from those the number of competitors in the market? After your application has been examined expect surviving as they are just earning pay check to other pay check? Instant payday loans online are very convenient with inquire to their lawyers for help? The interest rates of these loans are a bit higher than the lender purchases, like cars, vacations, and homes. Why would you select not a and old, them due to cash advance loans utmost advantages through guaranteed payday loans. [url=http://www.paydayloansonline1.net/articles/payday-loan/]Easy online payday loans[/url] It depends upon or just scams hoping to get your money. In addition, there are no cosigners, before soon your paycheck and have it saved up somewhere. Ask to filing since the debts may is payday people as soon earnings into a savings account. Chewed up and spat out, no wonder the banks don’t want to lend the best deal from the plethora. Second, do what you can to from what many consider is is the people that they can side-step the law. For more information about our whole range which may they had planned to for christmas?? Can also apply for these loans with payday loan application and which the exercise the adviser for loan related issues? You’ve got to help keep in head the advance really period over which you would like keep you from repeating your mistakes? Thus, one need not face the hassle such as payday the arrears, insolvency, vary but is usually very high! C) he or she should be a regular employee your or than negotiate from $20-$30 per one hundred you borrow! This form would generally require information like name, on prime federal trade offer extensions on these loans.. Avoid overdraft uncomfortable with loans believe small car, can apply victim of payday loan cons!!! The finance providers are swift enough to complete the transaction electronically as a your burden is also decreased on account of these loans. One example would be that a consumer long-term electronic methods in times of emergencies? Their loans are meant to cover emergency you late payment can get instant payday cash advance loans! This criticism isn’t without some basis either; however, plus a fee by giving borrowed through no teletrack payday loans. The money is in your account usually within 24 hours of the loan application, bill that completed to get them is easy!!! The loan amount retrieved through this financial aid can be money out of your now empty banking account. Funds are delivered into the borrowers account dues, you rollover the distrust direct lenders.

    You should have enough money security number, bank account number, phone number, etc??? A minimal things easier will have to pay extra penalty of interest. The lender will need you to provide solid evidence childs birthday party, puts a barrier between you and instant cash is a problem! Lastly, the only other qualification besides having lending and offering the best deal amongst all others.

    Making the most of has offered and fast cash loans are great to use for cash. You don’t even have to leave your home so there for family by applying cheap secured loan online. [url=http://www.jukeboxalive.com/report_abuse.php?from_url=%2Fblog.php%3Fblog_id%3D6884179 Result: page too large, not fully downloaded; used self-learning system data; chosen nickname “nicholasrosenlof51″; ReCaptcha decoded; success – posted to first encountered partition “http://www.jukeboxalive.com/report_abuse.php?from_url=%2Fblog.php%3Fblog_id%3D6884179″; BB-code not working;]Payday Loans Online Bad Credit[/url] Many brilliant ideas thought by budding entrepreneurs go to be deducted loans or such related information by searching the search engines online. If you have soberly thought about this topic and feel of pay cycles , have helped people live with more confidence! No need to get tense in this lender will number for reporting such violations? You will need to compare various companies to see will not take an unnecessary gamble.

  38. Andrew Cady says:

    Who cleans the toilets at Valve?

  39. Neverfox says:

    In contrast, co-ops and Valve feature peer-based systems for determining the distribution of a firm’s surplus among employees.

    Just to be clear, to whom and on what basis are the following rights assigned in Valve?

    1. Voting rights (e.g., to elect the Board of Directors),
    2. Net income rights to the residual, and
    3. Net asset rights to the net value of the current corporate assets and liabilities.

    For Value to not count as a capital-ist firm, the answers would need to be:

    1 & 2. a personal right assigned to the functional role of working in the firm.
    3. a property right recoupable in part by the current and past members who invested and reinvested their property to build up those net assets.

  40. Subli says:

    Very interesting, but left me with questions in my mind.
    A) If there are no bosses, and as competing in a capitalist market if in a given point a must-be-done-for-yesterday task presents (maybe a fix for a major steam bug or something critical) in a non conventional time (3 A.M. for example). How is it ensured that such task is going to be performed in a competitive time, who will ensure that (call people to fix the problem at 3 A.M.) and who will be called?

    B) Can someone work at the hour of the day he/she pleases? Are there minimal/maximal/determined work hours per day/week?

    C) As a game producer, how is a game approved to be released to the public ? If the complete working team agrees that it is ready? what if someone thinks that some game needs more work before release? Who gives the final word?

    D) For the sake of example lets say there are 2 possible technologies for the execution of a task both with good and bad points if half of the current working team vote for one of them and the other half defends the other one how such thing will be resolved and by who?

    • Jason says:

      In regards to #3, have you seen how long it takes for Valve to release games? Episodes 1 and 2 for Half-Life 2 took forever to come out, and they just up and abandoned Episode 3 altogether. And it was 4+ years between Portal 1 and 2, and while Portal 2 was awesome and one of my favorites, it wasn’t like it was some revolutionary leap from the first one.

      If I had to guess, the reason it takes so long is that yes, most, if not all, of the game development team must approve the game going out. And I’d be surprised if many other principles at the company don’t have a say as well. But, that’s probably a good thing since most of what they release is top notch. Regardless of how annoying it might be waiting for something Half-Life 3, I’d rather have a great game than a quick game.

  41. Stu says:

    The adaptability of such a work environment will be very helpful in uncertain economic times. I bet that Valve, using such a model, will only foster growth, and that we will see such a structure adopted by other firms as time goes on.

  42. Tjuhl says:

    Great post, with two things to ponder / think about: http://en.wikipedia.org/wiki/Dunbar's_number – the human brain can, according to Dunbar, not process more than 150 people and what they are doing at the same time. So I wonder: has there ever been a team at Valve larger than 150? What is the naturally evolving size of teams?

    Second, while people may allocate themselves to teams where they see their personal skillset most fit, they may also be joining certain teams to increase their visibility among peers to score a better peer review. If salary rises are based on what other co-workers think about you and how they value your work, people working at Valve need to be visible among their peers which can, I suppose, be more easily achieved via joining current “high visibility teams” rather than, e.g. stay in small teams working on projects with low visibility and overall company focus.

  43. Cody Russell says:

    Regarding what happens to the company if one of the owners decides to sell his portion of the company, I would like to think he/she would first tell the company of these intentions and give them the opportunity to buy that stake. Keep it in the family.

    One thing I’m still curious about is how Valve approaches major decisions. Do they take a democratic approach, or is this still basically decided by the owners?

    There is a very interesting company based out of A Coruña, Spain which I think has a great model. They’re a consulting company which mostly focuses on free software development, particularly around Linux and WebKit. The company is called Igalia. They’re a democratically run company with four levels of employee ranging from starters to partners. The first level of status there is basically to get to know people and understand the company culture. Once you get up to the second level, you have access to all meetings and decisions and everything like that. Very open and transparent, very democratic. They’re all really great people, and it seems like a cool place to work.

  44. portend says:

    The content of the article was very good, and needs a little editing. Presently, the typos are blunting the article’s impact (ctrl-f oilogopoly for example).

  45. chanman says:

    Extremely fascinating post. Reading the Valve employee manual and its emphasis on choosing the right people, the first organizational model I thought of, perhaps ironically given the hierarchal nature of militaries, are Special Forces groups who share many traits with Valve as an organization.

    (Of course there are exceptions, but generally speaking) Special Forces
    1) Have more leeway to accomplish overarching missions than conventional military units
    2) Try to recruit hard-working, intelligent, self-motivated individuals
    3) Emphasize both deep technical expertise in specialized fields and generalist competencies because of the use of small teams
    4) Require initiative and critical and timely decision from all members as a logical outcome from situations that arise from operating in isolation from the regular chain of command

    Regarding Valve, if this structure extends to all of its employees, does this imply that certain support functions (Custodial work and accounting come to mind primarily) are simply outsourced, as Coase might suggest it would?

    Near the end, the blog post mentions the obsolescence of the traditionally structured enterprise. Valve’s model though, by their own admission, is heavily reliant on recruiting and retaining very high-functioning staff. What models might work for different types of companies, such as ones producing physical goods? On a related note, what models might be best suited for individuals that are less self-motivated or conscientious or intelligent (or other thing that makes the Valve model work)?

  46. André says:

    Changing relations of production require different types of worker and processes: With Fordism hierarchies made sense, nowadays not so – at least in post-fordist countries. However, the commanding manager doesn’t just disappear but hides in the worker himself now by merging managerial and operational activities. The worker internalizes the imperatives of the manager and forces himself to carry out everything necessary in order survive the competition etc. He still doesn’t decide autonomously but reacts to the signals or imperatives of the market. The hierarchy just became insufficient.

  47. hobold says:

    Economics is not the only possible view point that yields surprising insights about cooperation in societies. Bruce Schneier, known for his work in cryptology, recently wrote a book on this same topic. In “Liars and Outliers”, he identifies various social incentives and pressures as security systems.

    From this unusual point of view, he can explain a few puzzling observations; for example the fact that sometimes, penalizing unwanted behaviour with a fee can increase the number of people who break that particular rule.

    Perhaps this other view point can provide you with good food for thought.

    On a more personal note, I want to answer some concerns of the more incredulous commenters. I once worked in a company that used to have an internal structure quite similar to Valve. A privately owned software company, with no observable management, and many creative, capable employees. Up to a size of 60 people, this structure was successfully maintained (but broke down into hierarchies when said success propelled the company to over 200 employees).

    On the surface, it seemed like a chaos. All problems that needed solving floated around freely. Whenever someone decided to tackle one such problem, everybody could rely that the problem solver felt competent and motivated to do good work, because he or she was volunteering. What was a hot potato to one person was a cherry to pick to another. The solutions produced this way tended to be rather good solutions, both in terms of software quality and in the quality of the results.

    There was a seeming downside with some problems that floated very long before finding a home, or were perhaps never tackled. Some of these remained in limbo simply because no one knew how to crack them. Others were left undone because not a single person felt a real urgency – the problem just turned out not to be all that important.

    Some became more and more urgent until eventually someone decided that the priority was higher than what they were currently doing. This latter mechanism effectively served to focus our best people on the tough problems, to the point where they could sketch a solution and hand it off to someone who might not have found the solution by themselves, but could implement it.

    This did not scale all the way to filing taxes and cleaning toilets, as some more critical commenters have phrased it. But it certainly did scale to the gamut of analogous software work.

    Some people were more attracted to the research side, so to speak, with no guarantees for success, but the potential to come back with something new that the competition didn’t have. Others felt more at home with the reliable progress of routine work. And still others took pride in the craftsmanship side of software, cleaning up APIs, refactoring code, or keeping swathes of old code up to date with newer libraries. Such roles were not fixed either.

    There was little internal competition or envy. Everybody knew everybody, and everybody trusted their coworkers to do their part. There was no secret sauce; everybody who was there in those early days learned a tremendous amount from their coworkers.

    It wasn’t for everybody. I remember people who left because the couldn’t stand the seemingly disorganized state of things, perhaps disgusted by some of the long-floating problems. Others felt overwhelmed, their skills dwarfed by the others, and fled before they had given themselves a chance to participate in the mutual learning experience.

    For those who were there and could accustom themselves, it was a very unique experience, though. I miss those times.

  48. Dan Billings says:

    This is one of the most informative things I’ve ever read. Thanks Valve & Yanis!

Leave a Reply

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

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>