The problem with Coase as I see it is it is a case of fitting a model to the facts, so has no predictive power at all, even if it is the correct explanation. Some of the factors are changing, as technology cuts information and transaction costs and makes markets cheaper, for example. So how much does that mean firms will shrink or grow?
Here are a few questions I wish yosefk.com would address:
1. Why do small companies exist? This is vice versa to his question, "Why do large firms exist?" -- if large companies have the advantage in eliminating transaction costs, small companies should be eliminated in favor of medium and large companies. (I'm taking into account Coase's "decreasing returns to the entrepreneur function.")
2. Is it possible small companies are more efficient at solving some problems -- and large companies others? For example, a large company should be better at, oh, say, contracting with the Department of Transportation for major freeway construction projects. I guess what I'm saying is: there may be an "endgame" where small companies dominate large companies (or the other way around) -- but in the here and now there's plenty of room for both. Evidenced by the fact that both types are still around, with no apparent trend eliminating either.
3. Are there any good reasons stated in the article or the Coase theorem that _don't_ involve intellectual property, for big companies to exist? (I can think of a few, but I want to hear what others have to say.) Intellectual property is so polarizing, and I'd prefer to think about Coase Theorem in different terms than it's phrased by yosefk.com.
I actually think the Coase Theorem does have the ability to predict economic outcomes, but remember that econ isn't usually about perfect predictions anyway. Econ is usually about identifying a better way to eliminate inefficiencies -- a knob to tweak, or a strategy, etc.
Even moving information between two teams in the same company has its costs.
The largest projects I've worked on (70-odd people) have worked best when everyone was working together in the same office. Matching that kind of efficiency with teams in different offices is impossible, let alone when they work for different companies.
Edit: Which isn't to say that outsourcing is impossible to implement well. But it needs to be things with well understood requirements so that little communication is needed. The more communication that is required, the harder it is to make it work well.
Good article - more people should be aware of Coase's work. However, I would surmise that PG was talking about "you" in the sense of "a hacker someone like myself", not about people in general, many of whom should not go try and create companies.
PG may have been talking to "a hacker someone like myself", but arbitrarily defining "you" is a bit of a fallacy in my opinion. If "you" can mean that, then why can't "you" refer just to the subset of people that could start successful companies?
Sorry for the semantics argument, but I don't really think PG (or anyone) can use "you" to refer anyone but whoever reads their work.
I think my previous company did not get the memo about us all being on the same team. You never knew who were gona stick a knife in your back. My personal theory some people do better playing the game at a big corp, some do better trying to run their own thing. Neither is inherently superior. Different games for different personalities.
I worked at a semiconductor mixed-signal startup - rife with control-loop theory, analog design, digital design/firmware, and lots of application support/software. They do exist. It's just very hard for them to last without eventually getting acquired by an established bigCo. But the same kind of rings true for a decent chunk of startups in the software space.
That said, most semiconductor startups do partnerships with bigger companies for second-sourcing or foundry help.
The transaction cost problem is true whether you're a startup or a big company. If anything it's better to be a highly-partnered startup than a startup-division at a big company as you may be able to avoid lots of the bureaucratic headaches the large company is accustomed to, helping to keep employees more motivated.
pg was specifically addressing the issue of management that is out of touch, while Coase was explaining the role of the firm. There is no reason why these reasons couldn't co-exist.
The main driver behind running a startup and working in a big company is a tension between two factors: control and leverage. As the chief of a small startup, a young person exerts a great deal of control over their destiny. However, they are short of resources. Even $20k from yc only goes so far. In contrast, at a bigger company, one has to constantly fight for resources. However, when they are available, and applied judiciously, the money can go a very long way. There is also the added benefit of working with some very brilliant technical people who may have no ounce of business-sense in their bones.
On the far end of the scale, when one has advanced further up the rank, there is a great deal of firepower at your disposal. It is actually the same with startups. No startup can forever stay at 2 employees. In the end, there has to be sensible management in place.
This isn't exactly criticism, but curiosity. I would like yosefk.com to address the following:
1. Is the revenue generated by a company proportionate to its size? Or do smaller companies make more? Or larger companies? Comparisons would be like WePay vs. Microsoft (or, in the same sector), WePay vs. PayPal.
2. Are the problems he notes in a hypothetical industry dominated by small companies outsourcing exclusive to hardware manufacturing, or to all aspects of the manufacturing process?
tl;dr -- Big companies are a good idea because they offer employment and suit the author's nature nicely because he possesses a strong "aversion to business."
Starting the article with pg's question--"Why be a manager when you could be a founder or early employee at a startup?"--then going on a different tangent altogether to answer "Why do large firms exist?" never left me satisfied with the author's claim he "could fill a book explaining why" to be a manager when you could be a founder or early employee at a startup. Claiming the question of why large firms exist to be the question implied by pg's original was a logical misstep.
pg's question appears--to me, so obviously I could be wrong--to dispute the wisdom of working at "companies where the only way to advance is to go into management." I'd have expected a "response" of this length to somehow defend either A) going to work for companies organized this way, 2) removing companies from the discussion altogether and defending the hierarchies & organizational principles behind management & its being the default rungs in the advancement ladder, or D) answer why being a manager is to be preferred vs the founder or early startup employee role.
The attempts to explain how markets work to the advantage of large firms--with random mentions of economic theory peppered about--does little to explain why they exist this way at all. It's merely a narrative of what exists now. A better--and more thorough and compelling--article would have tackled the big question of why our markets operate in such a way as to be of greater advantage to large firms who secure resources through contractual negotiations that evince preferential treatment to larger vs smaller buyers. If everyone interested in producing a product were able to achieve the same price-per-piece when buying two pieces as buying two million pieces, this shallow argument of why large firms exists--and are a good idea--would fall flat. The transaction costs the author uses as defense of the big company are artificial--an economic barrier placed atop the market that prohibits entry of competition by securing preferential treatment for one's own company.
Moreover, the article never gets back to the question it begins with--why work where going into management is the only way to advance? This is what I waited till the end of the article to find out. Quite disappointing. I'm still wondering. I've recently started working at a company where the only way to "advance" is to be in management. Before this, I was at a small company where I felt locked in as a programmer--where my ability to build reliable software was, of course, a benefit to the company they relied on, but where in 3 years I was never once asked to "advance" in a meaningful way that didn't mean becoming a "manager". I was pretty disappointed. Even in small startups, the business-as-usual notions of managers over programmers, designers, and other productive talent prevails. This company was better than others, but was eventually undone--a rift between the managers ultimately led them to shut down the entire company and lay everyone off. Now, having moved on to a larger company where I'm now "managing", I find myself wondering why this is any better of a lock-in than my previous post--meetings, emails, phone calls, reports, productivity ... all of this shit just bores me, gets in the way of getting a task done, and offers so little value to both my team's work & the company's bottom line. And of course here, as in so many other companies, advancement is tied to remuneration and other perks. But why should I put up with this shit instead of moving on? I work very hard to keep carving out time & tasks on which I can program directly, but it's tough.
The article touches on a couple key points that (I think) it erroneously attributes to being good points for big companies--namely, trust and cooperation. Moving from a small company to a larger company--and obviously, this can be anecdotal (but I don't think it can be so easily dismissed as such)--trust and cooperation are not abounding. Larger companies divide employees into distinct departments that, rather than operate in the company's best interests, operate in their manager's interest and draw up all kinds of political districts through which it seems impossible to build trust and a cooperative spirit among coworkers. There are departments who attempt to control parts of the business for which they have absolutely no skill, training, experience, or even basic understanding. Offering to work with them to help them meet their goals goes nowhere. Large firms do not equal a sudden conversion of competition to cooperation or mistrust to trust--which the author, to his credit, acknowledges. But neither do they truly "frame things right". Where questioning why to become a manager vs a founder or early employee, pg is questioning exactly what is fundamentally flawed--working in a place where advancement equals management (and, implied I think, where management is the only way to have a voice in the operations, visions, practices, and future of the company itself). Lack of trust and cooperation is a fundamental flaw of the market itself, which big companies both exacerbate and use in their favor. To be fair, there are undoubtedly certain companies which successfully create a culture of trust and cooperation among their employees--and these are companies likely to be admired and studied as examples for others. However, trust and cooperation must be both created and nurtured. Why can't the market do this, too? And are big companies at all at fault for the lack of trust and cooperation in the marketplace? Why would a large firm seek to increase overhead through hiring if they could establish trust and cooperation among partners in the marketplace?
I think the answer to the question requires a careful consideration of management itself--its goals and benefits, its responsibilities and rewards. This article adds little to that discussion. Furthermore, answering the question should entail a discussion of advancement itself. If we pull promotion to management out of the equation (which we should), what are we left with? What now qualifies as "advancement"? Is it just economic advancement? Is there advancement to positions of greater responsibility, impact, input, etc., that does not equal being turned into a manager? If a company (big or small) is hiring the right people for the right jobs, does anyone really need to be a manager? Should a manager's (or other "senior" person who is in a position to dictate what is to be done) idea of what to do next outweigh the input provided by a talented, experienced person who is actually going to do the productive work? Managers excel at creating a lot of fake work for employees and otherwise wasting everyone's time. This is, in my experience, overwhelmingly apparent and exacerbated in large firms--perhaps because small firms stay away from creating a culture of managers.
As it is, answering the question of "Why do large firms exist?" or the author's own premise of "When Big Companies are a Good Idea" needs far more than a marginally informative retelling of how large firms behave in the market today.
I'm going to throw out my opinion here (for discussion), but I should preface it by thanking you, I learned a lot from your post.
It seems large companies have a few distinct advantages:
1. They can operate divisions at a loss (e.g. Microsoft Entertainment and the XBox 1). They can "ride out" a down economy. This would be like having an angel with really deep pockets.
2. Stronger negotiating position in all cases: employees, clients, and vendors. Again, the investor network can compensate here for startups.
3. Kind of an outcome of #1 and #2: brand and leadership longevity. Sometimes you really can know more after having operated a business for decades.
My opinion is that all those advantages could create barriers to entry... but the big co's have enough problems of their own to let small co's innovate past them.
IMO this is good for both the large and the small, and especially good for the market: the established players can bring huge resources to bear on optimizing "solved" problems. The startups bring the right mix for getting ahead of the curve.
These are excellent points that get us closer to considering the question of when big companies are a good idea--especially #1. The ability to operate divisions at a loss and ride out a down economy is certainly an advantage had over the small firm who one might expect to be much harder hit when its core competency is hard pressed by external forces. So, one might be justified in saying a big company is a good idea when market forces bear negatively on participants, allowing the large firm to weather the storm less disastrously than a small firm. Of course, one can counter this proposition by arguing the large firm thus exhibits an unnaturally advantageous existence, not only bypassing but perhaps also eroding, perverting, retarding, and preventing evolution of the market in directions for which it is unprepared and in which it would lose its currently stronger standing. And yes, a similarly framed argument can be made for the large firm in the context of the evolution of markets--the larger, stronger company is the more fit, as evidenced by its ability to ride out market changes, retooling and repositioning itself for future success when the chaos dies down. However, I think it more likely this is only true as a result of its unfair advantage in having been large enough to not be naturally weakened & removed by the market as smaller firms are--because it is not removed from the market as a result of large concentrations of capital not possessed by smaller firms who are potentially better suited to adapt to changing conditions more rapidly but are forced out due to markets operating in line with extant large-firm bias (even when the market itself could benefit from the loss of the large firms).
#2 is, for me, quite close to the line of thinking offered in the OP, and of course the one I take issue with--it is more a statement of the way the market works now than a solid reason for why a big company is a good idea. I find this is a result of market players operating in such a manner as to give preferential treatment to a large company for reasons that don't actually equate to a larger company possessing this stronger negotiating position a priori. Rather, leveraging an existing preference that is itself an artificial construct (historically built by large firms) atop the market offers little in defense of why the large firm exists or is a good idea. In essence, the large firm enjoys a stronger negotiating position in the market because players in the market prefer to negotiate with a large firm because they believe it is in their best interest to do so because the large firm wields vastly larger capital with which it entices the players to negotiate with it over smaller players. This is circular and, in my view, ought then be discarded in attempts to explicate why this is a good idea.
#3 is a bit harder to pin down. I'm not convinced there is an inherent goodness in brand longevity as far as markets are concerned--or where employees, vendors, clients, or anyone else matters. I also find the material history of humanity and the markets replete with more examples of why leadership longevity is a bad thing than the reverse. There are certainly far more examples of long-term leaders ultimately driving their companies--large or small, it doesn't really matter--into relative oblivion or disarray or irrelevance than long-term leaders consistently innovating ahead of the markets, responding to changing conditions and rapidly correcting course so as to move forward. Leadership longevity often results in market players who have a vested interest in maintaining their position in the market through either coercion, prevention, or weathering out blips. It also can create companies who fail to change with the market because they believe they have it right and the market has it wrong. Perhaps in some cases this can be true ... but the perspective is always funny to me when people discuss "correcting" the market or some similar type of action. Correcting the market has always struck me as code for "get the market back in line with the operational goals of the large firms who have been benefitting the most." When the market moves against them, they seek correction and realignment. When the market moves toward them, they seek protection and cultivation.
Anyway, your points are far more interesting avenues to thoughtfully consider when big companies are a good idea. Not yet to the "Why do large firms exist?" question from the article yet, but at least getting closer to what I thought the author was going to do. And, we're still far from the question the author started the article with in the first place, of course.
