13 ms·

Beware Centralization

134 points by cinquemb 5y ago

jimkleiber 5y ago
What mainly catches my attention in this article is how it ends seemingly only implicating government for getting too large and not corporations. I like the argument that orgs can get too big and therefore not manage things well, and I feel confused why that would only apply to government and not to other orgs that control many aspects of our lives—e.g., Facebook has more users than any government in the world and access to human communication than most governments.
- cyanydeez 5y ago
  any org can turn sociopathic.
- clairity 5y ago
  one thing i thought it did a good job of was pointing out both corporations and governments suffer from the ills of consolidation.
  - jimkleiber 5y ago
    Oh yes, I think it did well on that, maybe that's why I felt so confused at the end, where the conclusion seemed to warn about government getting too big and not corporations, even tho most of the article seemed to warn about the ills of both. *Edit: fixed typo
    - sneak 5y ago
      The "ills" in the corporate sphere just result in reduced profits/efficiency. The "ills" of the central planning department of the government getting too large result in gulags.
      - jimkleiber 5y ago
        > The "ills" in the corporate sphere just result in reduced profits/efficiency. I disagree, as corporations can cause plenty of harm to people, whether physically or emotionally. > The "ills" of the central planning department of the government getting too large result in gulags. Yes, governments can have much more power over citizens, as moving from one to another can be much more difficult and the legitimacy to use violence to enforce, I agree. Maybe I just felt confused the original article didn't delve more into that, or maybe it did and I just didn't see the connection and it felt disjointed to me.
  - didibus 5y ago
    > one thing i thought it did a good job of was pointing out both corporations and governments suffer from the ills of consolidation Seriously? I saw nothing in the article to argue why consolidation would yield to such ills, it kind of just assumed it. Felt mostly like the article said: "why don't most firm consolidate? Well, maybe because it wouldn't end up so good? Maybe because the leader would fail to manage all of it properly?" I don't know, the whole thing was so thin on actual analysis, data, examples, annectodes, logic, etc. I felt it relied exclusively on the hope that you already are biased towards its conclusion to actually make its point. For example, it could have at the very least brought up very large successful firms that have consolidated? Like most tech giants? Or media corporations? And tried to explain why they have done so?
    - clairity 5y ago
      yes, it did a poor job of plainly spelling out the rationale, but not of putting gov and corp on equal footing here. (tech) giants might be successful, but the point is that they’re not as successful as they could be if they were loosely federated, more focused, independent companies. firms consolidate beyond the point of maximal economic benefit because of other reasons (at root, for power & status). the (lost) point being that this is suboptimal from a strictly economic perspective (same for overgrown governments).
      - didibus 5y ago
        > but the point is that they’re not as successful as they could be if they were loosely federated, more focused, independent companies I didn't see that point being made, I only saw it being affirmed without substantiated reasons. Why do you believe they aren't as successful in their consolidated state as they would be in a seperated state? What data, annectodes, rationale, experiments, experiences, etc. lead you to believe this to be a generalized truth? > but not of putting gov and corp on equal footing here Same thing, I saw nothing to justify why whatever may be true of corporations somehow would similarly apply to a different context such as governments. Why do you believe that what is generally true for a corporate model would be for a government? What data, annectodes, rationale, experiments, experiences, etc. lead you to believe this to be a generalized truth? And in both cases, are there exceptions to these truths? Are their circumstances that are required for them to hold? And if you're not going to walk me through the proof, at least make a case for why I should trust your assessment. Who are you and how much time and effort did you spend evaluating this? Did you do so in good faith? What did you do to make sure you challenged your pre-conceptions and didn't fall pray to bias? Did you have other people review your assessment? Were they themselves deep thinkers on the same topics? Do you have skin in the game? Did you test your theories in any way? Etc.
- bpodgursky 5y ago
  The problem is that Facebook (like IBM, etc) can fail under its own weight and incompetence without harming very many people. If the US government does... a lot of people get very badly hurt in the process.
- smolder 5y ago
  Not only does facebook have a tremendous size, amount of dysfunction, number of users, and pile of money, they have no responsibility to their users. There is no user representation at any $techco. The future of everything is being architected by closed door committee, free from competing ideas.
  - desertraven 5y ago
    As obvious as it is, I’d not considered this. Seems quite bizarre when the situation framed like so. Should some kind of democratic process apply to such organisations?
    - jimkleiber 5y ago
      I've started to experiment with looking at these large social media/communication orgs less as companies and more as countries/nation-states, and when I do, the metaphor illuminates a lot for me. E.g., FB gives its citizens free services as long as it can watch (almost) everything people do and put up advertisements (almost) everywhere. And if someone breaks the laws, the Supreme Court, appointed by the President, interprets those laws. If citizens don't like the laws, they can move to another country. And lastly, if citizens don't like the President, well, there's not much they can do, besides again, maybe move to another country. > Should some kind of democratic process apply to such organisations? I've wondered what would be some effective ways to add more user voice in the ownership and governance process of these platforms. Do you have any ideas?
      - bserge 5y ago
        They're also working on creating a New World Order™ so you can't easily move to another country, which, by the way, is a terribly underestimated and underused ability at the moment.
        jimkleiber 5y ago
        As the EU, a supranational government, strengthened in Europe, it actually made it easier to move to another country. Perhaps if the EU became more and more controlling of its member nations, then it wouldn't feel as if one were moving from another country, because each place would be quite similar in law, I don't agree that supranational governance would necessarily make it more difficult to move to another country. Also, I think it might be underused right now because it's probably really hard for most people around the world to move to another country for various reasons: financial, social, linguistic, legal, etc. I agree that moving countries can help people, I think it is currently a huge task for most and ironically, believe that better relationships between countries and an overarching form of governance may make it easier to do.
- wskinner 5y ago
  Because corporations are generally subject to competitive pressure, those that get so large they can no longer effectively function will go out of business. Government entities are in most cases legal monopolies and can continue to exist long after they become ineffective. See https://en.wikipedia.org/wiki/Creative_destruction https://en.wikipedia.org/wiki/Creative_destruction
  - JackFr 5y ago
    See also https://en.m.wikipedia.org/wiki/The_Nature_of_the_Firm https://en.m.wikipedia.org/wiki/The_Nature_of_the_Firm Ronald Coase’s paper from 1937 describing the pressures that lead to the formation of firms and those that constrain its size.
  - tshaddox 5y ago
    Unless of course they become so large that they are able to do thing to relieve themselves of competitive pressure.
    - jimkleiber 5y ago
      Exactly. I agree there are more competitive pressures for smaller orgs. Once some get really big, they can still face competitive pressure from smaller upstarts, that has happened many times. They can fold from being too big and not focused enough. They can fold from trying to stitch together too many cultures and have infighting that brings them down. They can also start to buy other smaller orgs to relieve competitive pressure. They can also create lock-in policies to make it very painful for people to leave. They can also spend money on political advertising to play on people's fears that certain individuals or orgs are out to get them. Government can also do this. I guess sometimes I just get confused why people seem to fear one and not both.
  - imtringued 5y ago
    The VC industry is intentionally overfeeding its startups to obtain monopolies. Returns on investment are so low the most attractive investment is a monopoly.
- sneak 5y ago
  Governments can legally employ violence to implement their policies and prerogatives. Corporations cannot.
  - jimkleiber 5y ago
    Yes, I agree that governments have more physical power over most people and can employ physical violence. I do think that corporations can also employ violence, albeit of an emotional, social, or even financial kind, against their users. If someone has a digital store and for some reason, Amazon or Shopify or Visa or Stripe or Paypal kicks them off their platform, that could ruin their business/livelihood/etc., which in turn could lead to suicide or homicide or drug abuse, etc. Again, I'm not saying losing one's online business is equivalent to being thrown in prison or many of the things governments can do to people, I'm just confused why people seem to be so afraid of the government getting too big and not also at least a little afraid of corporations getting too big.
    - fighterpilot 5y ago
      "I'm just confused why people seem to be so afraid of the government getting too big and not also at least a little afraid of corporations getting too big." You raise valid points but it remains the case that there isn't close to an equivalence there due to government's monopoly on force. All of the worst atrocities and dystopias, whether they've been inwards-focused (famines, hyperinflation, pogroms - North Korea, Venezuela, Cambodia, etc) or outwards-focused (invasions and genocides - Axis powers in WW2) have been perpetrated by governments. You could argue that these aren't real democracies, but liberal democracies are but a hop, skip and jump away from autocratic dictatorial hell as history has shown repeatedly. We have many modern mini-dystopias engineered by governments today. People are right to be fearful.
      - jimkleiber 5y ago
        I don't think all of the worst atrocities have been directly perpetuated by governments, with the most obvious one coming to mind being the transatlantic slave trade, where many companies facilitated the slave trade and also the slaveholding. Also gangs just came to mind. I do think that even if companies/non-governmental orgs are causing the violence, that ultimately we often look to government to resolve the conflict and/or to protect us, and if they don't, then it becomes their fault for not doing so. But again, overall, I'm not saying we shouldn't be afraid of governments wreaking absolute havoc and pain upon millions or billions of people, because, as you mention, "liberal democracies are but a hop, skip and jump away from autocratic dictatorial hell as history has shown repeatedly." Governments have harmed so many humans over the history of our species and I think it makes sense to be fearful they will get too powerful and do that. I just think it also makes sense to be a little afraid that corporations/other orgs will get large enough and powerful enough that they could also do damage to us.
- barsonme 5y ago
  Because Facebook doesn’t build roads or bridges. They can’t tax me or put me in jail. They’re can’t put migrant children in cages or drone strike weddings. They’re not responsible for making sure our water and air are clean. They can’t ban religions or sexuality. Facebook can’t Brexit. If Facebook fails, life goes on. If my government fails, life is going to suck. A lot.
  - jimkleiber 5y ago
    I guess I was not saying we should be equally afraid of corporations getting too centralized as we are of governments, I was just confused that it seemed to conclude with that we should only be afraid of governments getting too large. Large internet corporations can wreak havoc on one's digital life. Is that as bad as getting physically put in a cage or struck by a drone? I'd argue no. Can it still be deeply harmful to individuals? I'd argue yes. Maybe I misread the ending of the article and that the author meant that we should worry about the centralization of any org, and more so about the centralization of government orgs.
    - barsonme 5y ago
      > Maybe I misread the ending of the article and that the author meant that we should worry about the centralization of any org, and more so about the centralization of government orgs. fwiw, that’s what I got out of it.
      - jimkleiber 5y ago
        > These patterns, if true, are seem important regarding the ideal scales of both business and government. And I fear the U.S. public is insufficiently aware of them, as we seem to be on the verge of a historic increase in the scale and depth of government management of society. Maybe what hit me harder was the last line and the presumption that 1) it is a historic increase in the scale and depth of government management of society and 2) that we are also not seeing a large amount of corporate centralization. Maybe I perceive more corporate centralization than government centralization, as platforms have hundreds of millions or billions of users across 100+ countries, and I recall my 2005 political science class talking about the rising influence of multinational companies over nation-state governments.
- ByteJockey 5y ago
  The government is more dangerous than corporations. Mostly because of the size of the military.
  - femiagbabiaka 5y ago
    The military is made up of corporations now. From the Acela corridor to mercenaries who are making up large portions of boots on the ground troops.
    - ByteJockey 5y ago
      Y'know fair. I think defense contractors need to be considered separately from most corporations. They are inherently more dangerous than your garden variety corporation. Though, my understanding is that they are still not allowed to control nuclear weapons, so I'd still put them in sub-government levels of dangerous.
- Upazog 5y ago
  These things are like watching waves at the beach. You take a snapshot and show it to people and few will be able to say whether the wave is coming in or going out. Everyone will have an opinion one way or the other though. Thing to remember is, we are looking at dynamic ever changing system but discuss it in static snapshots. Thats always a source of confusion. Whats important about any centralization-decdntralization debate is that its a Tradeoff, ALWAYS a tradeoff, between Convenience and Freedom (See Tim Wu's book Master Switch for more). No one side ever wins because both have value in different situations. Look at covid as an example. Each govt comes out with its own vaccination, quarantine, treatment rules. This makes sense because they each have different capacities and resources. Singapore and Germany can do thing that Vietnam and Kenya cant. Now if my job involves moving silicon chips and their raw materials through 8 countries every day, all of a sudden my work load has jumped 8x cuz I have to follow 8 different sets of rules. I say screw that shit cuz my salary isnt jumping by 8x. Next thing you know factories are shutting down and people are being let go. If you have built up enough clout over the years, you can pull different govt officials into a room and say lets solve this, and what then emerges is more standardized rules. You could say that has centralized things that were decentralized. But thats not the point cuz tomorrow you know the Germans, who strictly follow the standards, realize they can be improved but why wait around for agreement from everyone, lets add the new ones and you can see how the cycle restarts. Things are always changing.
- Shmebulock 5y ago
  Because corporations die all the time. Governments don't
  - hannasanarion 5y ago
    Small, regional, and niche corporations die. Global conglomerates that are diversified into hundreds of different industries, many of which they have obtained near-absolute monopoly on, don't. Also, governments die all the time, there was a revolution in Afghanistan like three weeks ago, the new government hasn't even picked its flag yet, are you under a rock?
  - imtringued 5y ago
    Governments die all the time. We have elections for that.
- imtringued 5y ago
  It feels strange how the blog is called overcoming bias and yet the conclusion is heavily biased against governments. In my opinion the rise in "government intervention" can easily be explained by growing inequality and corporate power. People blame symptoms instead of causes. Zimbabwe clearly failed because they printed money, they clearly didn't fail because the agricultural reforms destroyed their export industries and made them dependent on food imports, no way.
  - jimkleiber 5y ago
    > It feels strange how the blog is called overcoming bias and yet the conclusion is heavily biased against governments. I appreciate this point and maybe that also contributed to my confusion at the conclusion which seemed pretty anti-government and inherently neutral or pro-business. > People blame symptoms instead of causes. The more I've thought about things over the last 1.5 years of covid and the Trump presidency, etc., I'm starting to think that we just blame people who we think are against us and make up stories and rationale to align with our "they're bad people" narrative. I think this often comes from a deeper, "I felt bad therefore this person made me feel bad therefore they are bad therefore they are the reason everything bad is happening in my life." And, as you mention, can cause us to focus on whichever symptom or belief of causality that aligns with that story, instead of overcoming our bias and trying to see how "good people" may have done "bad things" and "bad people" may have done "good things." Also, I appreciate your points about the Zimbabwean economy and hyper-inflation, I have a 10 trillion Zim note somewhere in my house and remember taking a class on the agricultural policies of Zim at that time.
- aww_dang 5y ago
  Most of these large corporations are large because of government interventions. Articles of incorporation are a legal concept granted by the state. The investment landscape is distorted by gov. interventions in the market. Access to investment capital, regulatory capture, Cantillon effects and other factors favor large corporations over the small. In a truly decentralized market environment, large players and monopolists would have greater difficulty competing with new entrants.
- pibechorro 5y ago
  Corporations are gov created entities. So its one and the same, especially after decades of regulatory capture and revolving doors of CEOs and officials. Its the same peoblem.
  - pibechorro 5y ago
    problem*
- csomar 5y ago
  You can opt out of Facebook. Opting out of your government is another deal, though. Also, your government control (to a large degree) Facebook and other organization. Government matter. Facebook is just another corporation.
clairity 5y ago
even as i nod along about the many ills of consolidation, i found the narrative not well-formed. it starts with a principle-agent type problem, and glides into a coordination type problem and lands somewhere around locus of control. these are all well-known/studied issues in management (and taught in mba programs). a little more structure and explicit elucidation of these issues would have made it better. i do heartily agree with the concluding paragraph that the general public seems largely unaware of these issues, to our collective detriment.
kdazzle 5y ago
Tldr; capitalism good because author “finds it obvious” in thought experiment
- TheGigaChad 5y ago
  Die commie rat.
ZhangSWEFAANG 5y ago
"Arguably the fact that there is the option to get at least the old profit levels, combined with many new options for making and using synergies across these firms, suggests that such merged firms can in general make more profits than they could separately." They never considered that you also introduce more risk.
- judge2020 5y ago
  The reward outweighs the risk since the chance of the risk coming to fruition at all is usually extremely small, even if it's dire when it happens.
desertraven 5y ago
The main concern seems to be that a centralised entity (corporate or government) can be flawed in its decision-making and/or execution. This is to the detriment of us all. What’s the solution here? Decentralisation doesn’t seem to compete well with centralised alternatives. Perhaps this will change with the latest wave of pro-decentralisation?
- cratermoon 5y ago
  > Decentralisation doesn’t seem to compete well with centralised alternatives Perhaps the problem is making competition the axle around which the world revolves.
  - jkhdigital 5y ago
    Competition (and capitalism) is not a conscious choice, it is simply what happens when humans are free to choose. Western societies consciously chose individual freedom, and the economic system simply followed directly from that choice.
    - cratermoon 5y ago
      Making competition the axle around which the world revolves is a choice.
    - cratermoon 5y ago
      Competition in general may not be a choice, but making it the axle around which the world revolves is a choice.
endisneigh 5y ago
All the flaws of centralization here also apply to decentralization tho
woodruffw 5y ago
Like with most of the posts on OB, this post makes a litany of confusing and incorrect assumptions: * That the forces present in the US's current agricultural system are not government-driven; * That the government and corporations are entities of a kind, and centralization in one can be used as a parable against the other without any further contextualization; * That we don't have ample, concrete evidence of the exact opposite outcomes predicted (i.e., developed, social democratic states that exercise central control over the food chain to the direct benefit of quality, safety, and public health).
- nicoffeine 5y ago
  When you don't have a good argument, it is best to stick with hypothetical situations. As soon as you move from straw men like "imagine three firms" to reality -- let's say, any F500 company -- it's too obvious that your hypothesis cannot be taken seriously. I don't understand why anyone would want a weak central government. Who's going to be there to bail out corporations during the next crisis?
  - shukantpal 5y ago
    Maybe corporations shouldn’t be bailed out.
cratermoon 5y ago
He makes a lot of proclamations about what would happen, but doesn't really back up his assertions. It's just a jumble of the usual drivel about "big government bad" restated as a priori facts.
- brink 5y ago
  It's why humanity moved away from monarchies and dictatorships to democracies and republics centuries ago; power corrupts. I think the author was hoping that it'd be common knowledge, but unfortunately it doesn't seem to be any more.
benatkin 5y ago
I was thinking that perhaps carbon in the atmosphere is a form of centralization. An externality of decentralization that causes centralization. And that sequestering carbon would be decentralizing it.
didibus 5y ago
I expected more from the article, it discusses an interesting question, but then it goes in a very light and superficial hypothetical analysis without much substance, too bad.
- cheschire 5y ago
  I expected more from this reply. It implies the existence of deep analysis but then just pontificates a bit. Too bad.
  - didibus 5y ago
    > It implies the existence of deep analysis All my comment implies is that I wish to read such deep analysis, and unfortunately the article didn't provide one. I'm hoping other people can reply with links to much better literature on the topic of consolidation versus seperation in business and politics.
vic_nyc 5y ago
Well, there is a name for this kind of centralization - it's called communism and it never worked.
xapata 5y ago
Thus the practical benefits of progressive taxation: it decentralizes.
alex_young 5y ago
Isn’t the US government especially good at centralization of specific industries? Education and mail service come to mind.
pphysch 5y ago
> But even in the absence of much corruption, civil servant selfishness, or partisan rancor, and even with the best political processes that we can imagine, a government managed food industry would still probably end up creating a worse world of food over the long run. Of course a perfect firm/government will perform better than the free market. That's why firms exist. Libertarian ideologues tend to have blissfully little knowledge or appreciation of the difficulties of scaling and effective management. Big firms sometimes underperform because scaling human organizations is a fucking hard problem, not because "centralization bad".
- TheGigaChad 5y ago
  Cocksucking corporate whore, get cancer and die.
- shukantpal 5y ago
  If a problem is fucking hard, then getting rid of it is better. Instead of projecting your arrogance into other ideologues, do some fucking homework.
  - pphysch 5y ago
    Intellectual cowardice--running away from challenges--is one way to describe libertarian ideologues, I'll give you that.
    - shukantpal 5y ago
      Oh yeah, ad hominem against libertarians is all you got. People like you shove their head into shit like pigs and think its a "challenge". Nice story bro.
dzink 5y ago
The key to governance is and always will be feedback. The human body is a great analog - millions of nerve endings, each of which can change the behavior of the whole system if it signals pain badly or persistently enough. Any system that provides low latency end points to its customers is preferred by customers - and small entrepreneurs and people highly invested in providing good service are the best end points so far. The larger the entity, the more layers need to be pushed between endings and decision maker, the higher the signal you need to emit to get heard. From an economic standpoint There are a bunch of downward pressure on corporate size: Capitalism doesn’t like companies that are too mangled with inefficient divisions - investors have portfolio goals and they can’t use a company in their portfolio that doesn’t signal like it’s industry. The more merged a company is, the more likely some divisions are to carry other divisions instead of returning money to stakeholders, the less attractive it becomes, the more likely their focused competitors are to get the big retirement fund allocation instead. Historically, Corporate Managers love few things more than more people under them, and lots of cash for toys on the budget. So in the 80s you had fat corporations with lots of cash at hand and private jetting. That triggered massive PE and leveraged buyout activity. Now companies know better than to stay cash ladden, because activist investors are hunting for fat corporations to LBO, especially with low interest rates out there. By this point, however, the fat is mostly gone. Centralization only helps with utilities, which is why if monopolies become too central to life in a country, they risk turning into one and then having profits capped. Feedback on quality is and always will be king and any giant is vulnerable to nimble and responsive competitors of higher quality.
banq 5y ago
Emergence: When things interact, they often birth new, unpredictable forms. Therefore, the sum total of a system is more than its component parts. As a system evolves, its structure can transform — just like how water becomes cold water until it turns into ice.
kristov 5y ago
I found the article odd. The stats of how many people work for companies greater than 250 can be accounted for without claiming companies get weaker as they bigger. Why do almost half of the population work for small companies then? Because that is how companies start. You don't create a new company with over 250 employees overnight. So your looking at a flow: companies start small and then grow, either by hiring more people or by getting swallowed up by a larger company. So if it's a flow then we should see huge megacorps periodically die? Well, a lot of company deaths are at the small end, where there is churn. Plus the system as a whole is growing - there is more money and "stuff" coming into existence every day. IBM still exists. The reason we don't have a ministry of food is not because such a ministry would be unstable as an organization (which seems to be the connection...). In fact such a ministry might be scarily stable, and produce food at a very low cost (efficiently). The reason we dont want a ministry of food is not related to efficiency or cost. Seems like the author just wanted to take a stab at the government. What's the takeaway here? We should have the water distribution system managed by thousands of tiny botique companies? City states with their own laws?
- jollybean 5y ago
  Countries that tend towards people in larger corps tend to be much richer. You need economies of scale to do a lot of things, and to compete effectively. There's a lot of profits in making cars, but Canada, Switzerland, Finland etc. even if they are relatively advanced countries, don't make cars. It's technically feasible, but really hard especially due to necessary scale and specialization.
jonahbenton 5y ago
For some reason my priors for Robin Hansen have been to anticipate intelligent, nuanced, insightful argument. And in practice every piece is the worst academic abstract banality that is also so plainly blind to what is going on in reality.
ShamblingMound 5y ago
The author asks: > So why don’t firms merge to get larger? But they do. Most industries/sectors trend toward consolidation. Specific examples that come to mind are radio, agriculture, banking, and tech. I did a quick search and found [0], which describes a general tendency toward consolidation as industries mature. One major reason for consolidation (ignored by the author) is economies of scale: as volume increases, the cost per unit decreases. On the other side, the author ignores benefits of central control for some things. For example, private railroad companies used to have different gauges for their railroads in order to protect their routes, but that led to unnecessary inefficiencies. Also, governments have a different mandate than the private sector. Governments are charged with pursuing citizens' values (freedom, equality, security, etc.), but for private sector companies, their goals boil down down maximizing profit. [0]: https://hbr.org/2002/12/the-consolidation-curve https://hbr.org/2002/12/the-consolidation-curve
- bumby 5y ago
  >Most industries/sectors trend toward consolidation. This reminded me of the discussion about how "mature" companies often struggle to innovate. The thought was that as companies mature, they tend to get larger and that size forces them to focus more on "maintenance" than innovation. So the way they "innovate" is by acquiring smaller companies (who are still in the innovation phase), resulting in eventual consolidation until a major disrupting event.
danguson 5y ago
It's hard when companies merge and have a new boss; prior CEOs' egos will not let this happen unless something is interesting that can improve the company's current state tremendously.
southerntofu 5y ago
> But this would be achieved in part via more uniformity, standardization, and stability of food processes. Government managed food would end up with less variety and adaptation to individual circumstances and preferences, and this food would improve and innovate less over time. Why would a public service for food have less diversity and innovation? A public service is operated by individuals, and doesn't have to be centralized or hierarchical. I agree with the point that reducing competition (by enabling cooperation) can be very efficient. But i strongly disagree it has to lead to centralization and therefore less options. If anything, capitalism leads to less choices because you need big money to even exist in a certain consumer space. There is currently no avenue for "garage innovators" to innovate and spread their products... a public service could (if not operated USSR/USA style) foster cooperation and innovation across the board.
jollybean 5y ago
It's interesting to read how tech people talk about industry in novel ways, but a lot of this material is well studied. A) The word 'centralization' is consistently misapplied by the tech crowd, in almost all cases, the term is 'monopoly' (or oligarchy or monopsony) and these are well understood and studied concepts in economics. B) The term 'Economies of Scale' is what he's inadvertently alluding to, but he's missing where most of the advantage of it comes from. Yes 'redundancy' is one artifact, but there are two bigger issues 1) specialization and 2) market power. When you're really big, you can start to afford to hire people to do things that competitors cannot, that might give you a big advantage. Most of real R&D falls into this category. So many applied 'AI' things are coming out of R&D divisions of big companies. 'Market Power' is a more abstract concept, but when you're large, you can make much stronger demands. Mondelez, P&G are big enough that they can plow through many competitors with all sorts of tactics: they get better prices and displays in-store, they have much bigger marketing budgets, they can sell at a loss in one corner to stifle a competitor, they can lean on buyers and distributors (i.e. buy them and you don't get our product) etc..You can think of Mondelez and P&G as 'owning' production and distribution. Apple is a 'Monopsony' for certain parts, and it has incredible leverage over it's supply chain and reaps most of the surplus profits from innovation in those supplying it. C) Many firms simply do not benefit from economies of scale i.e. roofers, carpenters, most kinds of law etc. but most do, however, the US has relatively speaking, much more people employed in 'big companies' that not - and that's what makes it rich. In fact this is a defining feature of 'Rich Europe' aka Germany, UK, Netherlands vs. 'Less Rich Europe' aka Poland, Spain: 'Rich Europe' has many more people in mid to larger companies. In many industries, you have to be at a certain scale to compete effectively. We generally don't like monopoly because the investors will tend to sit on an industry and extract rent as opposed to doing anything risky and actively block others from participating.
- mudita 5y ago
  I am not sure whom you mean, when you talk about "tech people”, but just in case you’re not aware: The author of the blog post is Robin Hanson, an associate professor of economics.
JetSetWilly 5y ago
Something like the NHS in the UK seems like a counterexample to this article's theory. A very large, extremely centralised state owned healthcare system, yet "health outcomes" seems perfectly in line with that of the intensely more expensive decentralised and privatised US system. And even in the recent pandemic, it was the NHS that discovered treatments like dexamethasone and tocilizumab thanks to an innovative and large scale A/B testing system that is much more difficult to set up in decentralised systems.
lottin 5y ago
The article talks about "concentration" not "centralisation". These are very different things. For example, a large organisation can be decentralised. This is not nitpicking, you can't think properly about stuff if you're using the wrong terminology.
xyzzy21 5y ago
Centralization also reduces redundancy which is essential to system reliability - a centralized system always ends up less reliable because there's more chance of single-point failure that cascades to all parts of the system quickly. This is especially true if "scale-free" features like "preferential attachment" exist. https://en.wikipedia.org/wiki/Preferential_attachment https://en.wikipedia.org/wiki/Preferential_attachment Consolidation of any type of thing (e.g. Wall Street and financial systems) can become very fragile and many of the guards against failure (risk models) cease to be valid the moment you have preferential attachment.
mark_l_watson 5y ago
I largely agree with the author, but they left out something important: arguably the most important thing that large corporations do to protect their interests is co-opting the government with campaign contributions, promises of future lucrative private sector employment, etc. Combined with control of the news media to divert attention, huge corporations have superpowers compared to small businesses and individuals.
zestyping 5y ago
This might be the sloppiest take I've ever seen on this topic. - "In fact firms do not do this." — In fact firms merge all the time. - "The typical scale of most firms seems far smaller than can be explained by these effects." — No reason is given. Italics do not constitute an argument. - "It seems obvious to me that by now such a food agency would have intervened in food production, processing, and distribution far more extensively that has been the actual case in our actual history." — Not at all obvious, and again no evidence or even any argument is given. - "The amazing thing is that all this would happen even with high quality oversight and accountability by agencies to politicians, and politicians to voters." — The amazing thing is that the author thinks he can state this prediction without basis of any sort. There are good reasons to beware centralization, but this piece is simply lazy. This gets an F. Show your work.

134 points by cinquemb 5y ago

What mainly catches my attention in this article is how it ends seemingly only implicating government for getting too large and not corporations. I like the argument that orgs can get too big and therefore not manage things well, and I feel confused why that would only apply to government and not to other orgs that control many aspects of our lives—e.g., Facebook has more users than any government in the world and access to human communication than most governments.

cyanydeez 5y ago
any org can turn sociopathic.
clairity 5y ago
one thing i thought it did a good job of was pointing out both corporations and governments suffer from the ills of consolidation.
- jimkleiber 5y ago
  Oh yes, I think it did well on that, maybe that's why I felt so confused at the end, where the conclusion seemed to warn about government getting too big and not corporations, even tho most of the article seemed to warn about the ills of both. *Edit: fixed typo
  - sneak 5y ago
    The "ills" in the corporate sphere just result in reduced profits/efficiency. The "ills" of the central planning department of the government getting too large result in gulags.
    - jimkleiber 5y ago
      > The "ills" in the corporate sphere just result in reduced profits/efficiency. I disagree, as corporations can cause plenty of harm to people, whether physically or emotionally. > The "ills" of the central planning department of the government getting too large result in gulags. Yes, governments can have much more power over citizens, as moving from one to another can be much more difficult and the legitimacy to use violence to enforce, I agree. Maybe I just felt confused the original article didn't delve more into that, or maybe it did and I just didn't see the connection and it felt disjointed to me.
- didibus 5y ago
  > one thing i thought it did a good job of was pointing out both corporations and governments suffer from the ills of consolidation Seriously? I saw nothing in the article to argue why consolidation would yield to such ills, it kind of just assumed it. Felt mostly like the article said: "why don't most firm consolidate? Well, maybe because it wouldn't end up so good? Maybe because the leader would fail to manage all of it properly?" I don't know, the whole thing was so thin on actual analysis, data, examples, annectodes, logic, etc. I felt it relied exclusively on the hope that you already are biased towards its conclusion to actually make its point. For example, it could have at the very least brought up very large successful firms that have consolidated? Like most tech giants? Or media corporations? And tried to explain why they have done so?
  - clairity 5y ago
    yes, it did a poor job of plainly spelling out the rationale, but not of putting gov and corp on equal footing here. (tech) giants might be successful, but the point is that they’re not as successful as they could be if they were loosely federated, more focused, independent companies. firms consolidate beyond the point of maximal economic benefit because of other reasons (at root, for power & status). the (lost) point being that this is suboptimal from a strictly economic perspective (same for overgrown governments).
    - didibus 5y ago
      > but the point is that they’re not as successful as they could be if they were loosely federated, more focused, independent companies I didn't see that point being made, I only saw it being affirmed without substantiated reasons. Why do you believe they aren't as successful in their consolidated state as they would be in a seperated state? What data, annectodes, rationale, experiments, experiences, etc. lead you to believe this to be a generalized truth? > but not of putting gov and corp on equal footing here Same thing, I saw nothing to justify why whatever may be true of corporations somehow would similarly apply to a different context such as governments. Why do you believe that what is generally true for a corporate model would be for a government? What data, annectodes, rationale, experiments, experiences, etc. lead you to believe this to be a generalized truth? And in both cases, are there exceptions to these truths? Are their circumstances that are required for them to hold? And if you're not going to walk me through the proof, at least make a case for why I should trust your assessment. Who are you and how much time and effort did you spend evaluating this? Did you do so in good faith? What did you do to make sure you challenged your pre-conceptions and didn't fall pray to bias? Did you have other people review your assessment? Were they themselves deep thinkers on the same topics? Do you have skin in the game? Did you test your theories in any way? Etc.
bpodgursky 5y ago
The problem is that Facebook (like IBM, etc) can fail under its own weight and incompetence without harming very many people. If the US government does... a lot of people get very badly hurt in the process.
smolder 5y ago
Not only does facebook have a tremendous size, amount of dysfunction, number of users, and pile of money, they have no responsibility to their users. There is no user representation at any $techco. The future of everything is being architected by closed door committee, free from competing ideas.
- desertraven 5y ago
  As obvious as it is, I’d not considered this. Seems quite bizarre when the situation framed like so. Should some kind of democratic process apply to such organisations?
  - jimkleiber 5y ago
    I've started to experiment with looking at these large social media/communication orgs less as companies and more as countries/nation-states, and when I do, the metaphor illuminates a lot for me. E.g., FB gives its citizens free services as long as it can watch (almost) everything people do and put up advertisements (almost) everywhere. And if someone breaks the laws, the Supreme Court, appointed by the President, interprets those laws. If citizens don't like the laws, they can move to another country. And lastly, if citizens don't like the President, well, there's not much they can do, besides again, maybe move to another country. > Should some kind of democratic process apply to such organisations? I've wondered what would be some effective ways to add more user voice in the ownership and governance process of these platforms. Do you have any ideas?
    - bserge 5y ago
      They're also working on creating a New World Order™ so you can't easily move to another country, which, by the way, is a terribly underestimated and underused ability at the moment.
      - jimkleiber 5y ago
        As the EU, a supranational government, strengthened in Europe, it actually made it easier to move to another country. Perhaps if the EU became more and more controlling of its member nations, then it wouldn't feel as if one were moving from another country, because each place would be quite similar in law, I don't agree that supranational governance would necessarily make it more difficult to move to another country. Also, I think it might be underused right now because it's probably really hard for most people around the world to move to another country for various reasons: financial, social, linguistic, legal, etc. I agree that moving countries can help people, I think it is currently a huge task for most and ironically, believe that better relationships between countries and an overarching form of governance may make it easier to do.
wskinner 5y ago
Because corporations are generally subject to competitive pressure, those that get so large they can no longer effectively function will go out of business. Government entities are in most cases legal monopolies and can continue to exist long after they become ineffective. See https://en.wikipedia.org/wiki/Creative_destruction https://en.wikipedia.org/wiki/Creative_destruction
- JackFr 5y ago
  See also https://en.m.wikipedia.org/wiki/The_Nature_of_the_Firm https://en.m.wikipedia.org/wiki/The_Nature_of_the_Firm Ronald Coase’s paper from 1937 describing the pressures that lead to the formation of firms and those that constrain its size.
- tshaddox 5y ago
  Unless of course they become so large that they are able to do thing to relieve themselves of competitive pressure.
  - jimkleiber 5y ago
    Exactly. I agree there are more competitive pressures for smaller orgs. Once some get really big, they can still face competitive pressure from smaller upstarts, that has happened many times. They can fold from being too big and not focused enough. They can fold from trying to stitch together too many cultures and have infighting that brings them down. They can also start to buy other smaller orgs to relieve competitive pressure. They can also create lock-in policies to make it very painful for people to leave. They can also spend money on political advertising to play on people's fears that certain individuals or orgs are out to get them. Government can also do this. I guess sometimes I just get confused why people seem to fear one and not both.
- imtringued 5y ago
  The VC industry is intentionally overfeeding its startups to obtain monopolies. Returns on investment are so low the most attractive investment is a monopoly.
sneak 5y ago
Governments can legally employ violence to implement their policies and prerogatives. Corporations cannot.
- jimkleiber 5y ago
  Yes, I agree that governments have more physical power over most people and can employ physical violence. I do think that corporations can also employ violence, albeit of an emotional, social, or even financial kind, against their users. If someone has a digital store and for some reason, Amazon or Shopify or Visa or Stripe or Paypal kicks them off their platform, that could ruin their business/livelihood/etc., which in turn could lead to suicide or homicide or drug abuse, etc. Again, I'm not saying losing one's online business is equivalent to being thrown in prison or many of the things governments can do to people, I'm just confused why people seem to be so afraid of the government getting too big and not also at least a little afraid of corporations getting too big.
  - fighterpilot 5y ago
    "I'm just confused why people seem to be so afraid of the government getting too big and not also at least a little afraid of corporations getting too big." You raise valid points but it remains the case that there isn't close to an equivalence there due to government's monopoly on force. All of the worst atrocities and dystopias, whether they've been inwards-focused (famines, hyperinflation, pogroms - North Korea, Venezuela, Cambodia, etc) or outwards-focused (invasions and genocides - Axis powers in WW2) have been perpetrated by governments. You could argue that these aren't real democracies, but liberal democracies are but a hop, skip and jump away from autocratic dictatorial hell as history has shown repeatedly. We have many modern mini-dystopias engineered by governments today. People are right to be fearful.
    - jimkleiber 5y ago
      I don't think all of the worst atrocities have been directly perpetuated by governments, with the most obvious one coming to mind being the transatlantic slave trade, where many companies facilitated the slave trade and also the slaveholding. Also gangs just came to mind. I do think that even if companies/non-governmental orgs are causing the violence, that ultimately we often look to government to resolve the conflict and/or to protect us, and if they don't, then it becomes their fault for not doing so. But again, overall, I'm not saying we shouldn't be afraid of governments wreaking absolute havoc and pain upon millions or billions of people, because, as you mention, "liberal democracies are but a hop, skip and jump away from autocratic dictatorial hell as history has shown repeatedly." Governments have harmed so many humans over the history of our species and I think it makes sense to be fearful they will get too powerful and do that. I just think it also makes sense to be a little afraid that corporations/other orgs will get large enough and powerful enough that they could also do damage to us.
barsonme 5y ago
Because Facebook doesn’t build roads or bridges. They can’t tax me or put me in jail. They’re can’t put migrant children in cages or drone strike weddings. They’re not responsible for making sure our water and air are clean. They can’t ban religions or sexuality. Facebook can’t Brexit. If Facebook fails, life goes on. If my government fails, life is going to suck. A lot.
- jimkleiber 5y ago
  I guess I was not saying we should be equally afraid of corporations getting too centralized as we are of governments, I was just confused that it seemed to conclude with that we should only be afraid of governments getting too large. Large internet corporations can wreak havoc on one's digital life. Is that as bad as getting physically put in a cage or struck by a drone? I'd argue no. Can it still be deeply harmful to individuals? I'd argue yes. Maybe I misread the ending of the article and that the author meant that we should worry about the centralization of any org, and more so about the centralization of government orgs.
  - barsonme 5y ago
    > Maybe I misread the ending of the article and that the author meant that we should worry about the centralization of any org, and more so about the centralization of government orgs. fwiw, that’s what I got out of it.
    - jimkleiber 5y ago
      > These patterns, if true, are seem important regarding the ideal scales of both business and government. And I fear the U.S. public is insufficiently aware of them, as we seem to be on the verge of a historic increase in the scale and depth of government management of society. Maybe what hit me harder was the last line and the presumption that 1) it is a historic increase in the scale and depth of government management of society and 2) that we are also not seeing a large amount of corporate centralization. Maybe I perceive more corporate centralization than government centralization, as platforms have hundreds of millions or billions of users across 100+ countries, and I recall my 2005 political science class talking about the rising influence of multinational companies over nation-state governments.
ByteJockey 5y ago
The government is more dangerous than corporations. Mostly because of the size of the military.
- femiagbabiaka 5y ago
  The military is made up of corporations now. From the Acela corridor to mercenaries who are making up large portions of boots on the ground troops.
  - ByteJockey 5y ago
    Y'know fair. I think defense contractors need to be considered separately from most corporations. They are inherently more dangerous than your garden variety corporation. Though, my understanding is that they are still not allowed to control nuclear weapons, so I'd still put them in sub-government levels of dangerous.
Upazog 5y ago
These things are like watching waves at the beach. You take a snapshot and show it to people and few will be able to say whether the wave is coming in or going out. Everyone will have an opinion one way or the other though. Thing to remember is, we are looking at dynamic ever changing system but discuss it in static snapshots. Thats always a source of confusion. Whats important about any centralization-decdntralization debate is that its a Tradeoff, ALWAYS a tradeoff, between Convenience and Freedom (See Tim Wu's book Master Switch for more). No one side ever wins because both have value in different situations. Look at covid as an example. Each govt comes out with its own vaccination, quarantine, treatment rules. This makes sense because they each have different capacities and resources. Singapore and Germany can do thing that Vietnam and Kenya cant. Now if my job involves moving silicon chips and their raw materials through 8 countries every day, all of a sudden my work load has jumped 8x cuz I have to follow 8 different sets of rules. I say screw that shit cuz my salary isnt jumping by 8x. Next thing you know factories are shutting down and people are being let go. If you have built up enough clout over the years, you can pull different govt officials into a room and say lets solve this, and what then emerges is more standardized rules. You could say that has centralized things that were decentralized. But thats not the point cuz tomorrow you know the Germans, who strictly follow the standards, realize they can be improved but why wait around for agreement from everyone, lets add the new ones and you can see how the cycle restarts. Things are always changing.
Shmebulock 5y ago
Because corporations die all the time. Governments don't
- hannasanarion 5y ago
  Small, regional, and niche corporations die. Global conglomerates that are diversified into hundreds of different industries, many of which they have obtained near-absolute monopoly on, don't. Also, governments die all the time, there was a revolution in Afghanistan like three weeks ago, the new government hasn't even picked its flag yet, are you under a rock?
- imtringued 5y ago
  Governments die all the time. We have elections for that.
imtringued 5y ago
It feels strange how the blog is called overcoming bias and yet the conclusion is heavily biased against governments. In my opinion the rise in "government intervention" can easily be explained by growing inequality and corporate power. People blame symptoms instead of causes. Zimbabwe clearly failed because they printed money, they clearly didn't fail because the agricultural reforms destroyed their export industries and made them dependent on food imports, no way.
- jimkleiber 5y ago
  > It feels strange how the blog is called overcoming bias and yet the conclusion is heavily biased against governments. I appreciate this point and maybe that also contributed to my confusion at the conclusion which seemed pretty anti-government and inherently neutral or pro-business. > People blame symptoms instead of causes. The more I've thought about things over the last 1.5 years of covid and the Trump presidency, etc., I'm starting to think that we just blame people who we think are against us and make up stories and rationale to align with our "they're bad people" narrative. I think this often comes from a deeper, "I felt bad therefore this person made me feel bad therefore they are bad therefore they are the reason everything bad is happening in my life." And, as you mention, can cause us to focus on whichever symptom or belief of causality that aligns with that story, instead of overcoming our bias and trying to see how "good people" may have done "bad things" and "bad people" may have done "good things." Also, I appreciate your points about the Zimbabwean economy and hyper-inflation, I have a 10 trillion Zim note somewhere in my house and remember taking a class on the agricultural policies of Zim at that time.
aww_dang 5y ago
Most of these large corporations are large because of government interventions. Articles of incorporation are a legal concept granted by the state. The investment landscape is distorted by gov. interventions in the market. Access to investment capital, regulatory capture, Cantillon effects and other factors favor large corporations over the small. In a truly decentralized market environment, large players and monopolists would have greater difficulty competing with new entrants.
pibechorro 5y ago
Corporations are gov created entities. So its one and the same, especially after decades of regulatory capture and revolving doors of CEOs and officials. Its the same peoblem.
- pibechorro 5y ago
  problem*
csomar 5y ago
You can opt out of Facebook. Opting out of your government is another deal, though. Also, your government control (to a large degree) Facebook and other organization. Government matter. Facebook is just another corporation.

clairity 5y ago

even as i nod along about the many ills of consolidation, i found the narrative not well-formed. it starts with a principle-agent type problem, and glides into a coordination type problem and lands somewhere around locus of control. these are all well-known/studied issues in management (and taught in mba programs). a little more structure and explicit elucidation of these issues would have made it better. i do heartily agree with the concluding paragraph that the general public seems largely unaware of these issues, to our collective detriment.

kdazzle 5y ago

Tldr; capitalism good because author “finds it obvious” in thought experiment

TheGigaChad 5y ago
Die commie rat.

ZhangSWEFAANG 5y ago

"Arguably the fact that there is the option to get at least the old profit levels, combined with many new options for making and using synergies across these firms, suggests that such merged firms can in general make more profits than they could separately." They never considered that you also introduce more risk.

judge2020 5y ago
The reward outweighs the risk since the chance of the risk coming to fruition at all is usually extremely small, even if it's dire when it happens.

desertraven 5y ago

The main concern seems to be that a centralised entity (corporate or government) can be flawed in its decision-making and/or execution. This is to the detriment of us all. What’s the solution here? Decentralisation doesn’t seem to compete well with centralised alternatives. Perhaps this will change with the latest wave of pro-decentralisation?

cratermoon 5y ago
> Decentralisation doesn’t seem to compete well with centralised alternatives Perhaps the problem is making competition the axle around which the world revolves.
- jkhdigital 5y ago
  Competition (and capitalism) is not a conscious choice, it is simply what happens when humans are free to choose. Western societies consciously chose individual freedom, and the economic system simply followed directly from that choice.
  - cratermoon 5y ago
    Making competition the axle around which the world revolves is a choice.
  - cratermoon 5y ago
    Competition in general may not be a choice, but making it the axle around which the world revolves is a choice.

endisneigh 5y ago

All the flaws of centralization here also apply to decentralization tho

woodruffw 5y ago

Like with most of the posts on OB, this post makes a litany of confusing and incorrect assumptions: * That the forces present in the US's current agricultural system are not government-driven; * That the government and corporations are entities of a kind, and centralization in one can be used as a parable against the other without any further contextualization; * That we don't have ample, concrete evidence of the exact opposite outcomes predicted (i.e., developed, social democratic states that exercise central control over the food chain to the direct benefit of quality, safety, and public health).

nicoffeine 5y ago
When you don't have a good argument, it is best to stick with hypothetical situations. As soon as you move from straw men like "imagine three firms" to reality -- let's say, any F500 company -- it's too obvious that your hypothesis cannot be taken seriously. I don't understand why anyone would want a weak central government. Who's going to be there to bail out corporations during the next crisis?
- shukantpal 5y ago
  Maybe corporations shouldn’t be bailed out.

cratermoon 5y ago

He makes a lot of proclamations about what would happen, but doesn't really back up his assertions. It's just a jumble of the usual drivel about "big government bad" restated as a priori facts.

brink 5y ago
It's why humanity moved away from monarchies and dictatorships to democracies and republics centuries ago; power corrupts. I think the author was hoping that it'd be common knowledge, but unfortunately it doesn't seem to be any more.

benatkin 5y ago

I was thinking that perhaps carbon in the atmosphere is a form of centralization. An externality of decentralization that causes centralization. And that sequestering carbon would be decentralizing it.

didibus 5y ago

I expected more from the article, it discusses an interesting question, but then it goes in a very light and superficial hypothetical analysis without much substance, too bad.

cheschire 5y ago
I expected more from this reply. It implies the existence of deep analysis but then just pontificates a bit. Too bad.
- didibus 5y ago
  > It implies the existence of deep analysis All my comment implies is that I wish to read such deep analysis, and unfortunately the article didn't provide one. I'm hoping other people can reply with links to much better literature on the topic of consolidation versus seperation in business and politics.

vic_nyc 5y ago

Well, there is a name for this kind of centralization - it's called communism and it never worked.

xapata 5y ago

Thus the practical benefits of progressive taxation: it decentralizes.

alex_young 5y ago

Isn’t the US government especially good at centralization of specific industries? Education and mail service come to mind.

pphysch 5y ago

> But even in the absence of much corruption, civil servant selfishness, or partisan rancor, and even with the best political processes that we can imagine, a government managed food industry would still probably end up creating a worse world of food over the long run. Of course a perfect firm/government will perform better than the free market. That's why firms exist. Libertarian ideologues tend to have blissfully little knowledge or appreciation of the difficulties of scaling and effective management. Big firms sometimes underperform because scaling human organizations is a fucking hard problem, not because "centralization bad".

TheGigaChad 5y ago
Cocksucking corporate whore, get cancer and die.
shukantpal 5y ago
If a problem is fucking hard, then getting rid of it is better. Instead of projecting your arrogance into other ideologues, do some fucking homework.
- pphysch 5y ago
  Intellectual cowardice--running away from challenges--is one way to describe libertarian ideologues, I'll give you that.
  - shukantpal 5y ago
    Oh yeah, ad hominem against libertarians is all you got. People like you shove their head into shit like pigs and think its a "challenge". Nice story bro.

dzink 5y ago

The key to governance is and always will be feedback. The human body is a great analog - millions of nerve endings, each of which can change the behavior of the whole system if it signals pain badly or persistently enough. Any system that provides low latency end points to its customers is preferred by customers - and small entrepreneurs and people highly invested in providing good service are the best end points so far. The larger the entity, the more layers need to be pushed between endings and decision maker, the higher the signal you need to emit to get heard. From an economic standpoint There are a bunch of downward pressure on corporate size: Capitalism doesn’t like companies that are too mangled with inefficient divisions - investors have portfolio goals and they can’t use a company in their portfolio that doesn’t signal like it’s industry. The more merged a company is, the more likely some divisions are to carry other divisions instead of returning money to stakeholders, the less attractive it becomes, the more likely their focused competitors are to get the big retirement fund allocation instead. Historically, Corporate Managers love few things more than more people under them, and lots of cash for toys on the budget. So in the 80s you had fat corporations with lots of cash at hand and private jetting. That triggered massive PE and leveraged buyout activity. Now companies know better than to stay cash ladden, because activist investors are hunting for fat corporations to LBO, especially with low interest rates out there. By this point, however, the fat is mostly gone. Centralization only helps with utilities, which is why if monopolies become too central to life in a country, they risk turning into one and then having profits capped. Feedback on quality is and always will be king and any giant is vulnerable to nimble and responsive competitors of higher quality.

banq 5y ago

Emergence: When things interact, they often birth new, unpredictable forms. Therefore, the sum total of a system is more than its component parts. As a system evolves, its structure can transform — just like how water becomes cold water until it turns into ice.

kristov 5y ago

I found the article odd. The stats of how many people work for companies greater than 250 can be accounted for without claiming companies get weaker as they bigger. Why do almost half of the population work for small companies then? Because that is how companies start. You don't create a new company with over 250 employees overnight. So your looking at a flow: companies start small and then grow, either by hiring more people or by getting swallowed up by a larger company. So if it's a flow then we should see huge megacorps periodically die? Well, a lot of company deaths are at the small end, where there is churn. Plus the system as a whole is growing - there is more money and "stuff" coming into existence every day. IBM still exists. The reason we don't have a ministry of food is not because such a ministry would be unstable as an organization (which seems to be the connection...). In fact such a ministry might be scarily stable, and produce food at a very low cost (efficiently). The reason we dont want a ministry of food is not related to efficiency or cost. Seems like the author just wanted to take a stab at the government. What's the takeaway here? We should have the water distribution system managed by thousands of tiny botique companies? City states with their own laws?

jollybean 5y ago
Countries that tend towards people in larger corps tend to be much richer. You need economies of scale to do a lot of things, and to compete effectively. There's a lot of profits in making cars, but Canada, Switzerland, Finland etc. even if they are relatively advanced countries, don't make cars. It's technically feasible, but really hard especially due to necessary scale and specialization.

jonahbenton 5y ago

For some reason my priors for Robin Hansen have been to anticipate intelligent, nuanced, insightful argument. And in practice every piece is the worst academic abstract banality that is also so plainly blind to what is going on in reality.

ShamblingMound 5y ago

The author asks: > So why don’t firms merge to get larger? But they do. Most industries/sectors trend toward consolidation. Specific examples that come to mind are radio, agriculture, banking, and tech. I did a quick search and found [0], which describes a general tendency toward consolidation as industries mature. One major reason for consolidation (ignored by the author) is economies of scale: as volume increases, the cost per unit decreases. On the other side, the author ignores benefits of central control for some things. For example, private railroad companies used to have different gauges for their railroads in order to protect their routes, but that led to unnecessary inefficiencies. Also, governments have a different mandate than the private sector. Governments are charged with pursuing citizens' values (freedom, equality, security, etc.), but for private sector companies, their goals boil down down maximizing profit. [0]: https://hbr.org/2002/12/the-consolidation-curve https://hbr.org/2002/12/the-consolidation-curve

bumby 5y ago
>Most industries/sectors trend toward consolidation. This reminded me of the discussion about how "mature" companies often struggle to innovate. The thought was that as companies mature, they tend to get larger and that size forces them to focus more on "maintenance" than innovation. So the way they "innovate" is by acquiring smaller companies (who are still in the innovation phase), resulting in eventual consolidation until a major disrupting event.

danguson 5y ago

It's hard when companies merge and have a new boss; prior CEOs' egos will not let this happen unless something is interesting that can improve the company's current state tremendously.

southerntofu 5y ago

> But this would be achieved in part via more uniformity, standardization, and stability of food processes. Government managed food would end up with less variety and adaptation to individual circumstances and preferences, and this food would improve and innovate less over time. Why would a public service for food have less diversity and innovation? A public service is operated by individuals, and doesn't have to be centralized or hierarchical. I agree with the point that reducing competition (by enabling cooperation) can be very efficient. But i strongly disagree it has to lead to centralization and therefore less options. If anything, capitalism leads to less choices because you need big money to even exist in a certain consumer space. There is currently no avenue for "garage innovators" to innovate and spread their products... a public service could (if not operated USSR/USA style) foster cooperation and innovation across the board.

jollybean 5y ago

It's interesting to read how tech people talk about industry in novel ways, but a lot of this material is well studied. A) The word 'centralization' is consistently misapplied by the tech crowd, in almost all cases, the term is 'monopoly' (or oligarchy or monopsony) and these are well understood and studied concepts in economics. B) The term 'Economies of Scale' is what he's inadvertently alluding to, but he's missing where most of the advantage of it comes from. Yes 'redundancy' is one artifact, but there are two bigger issues 1) specialization and 2) market power. When you're really big, you can start to afford to hire people to do things that competitors cannot, that might give you a big advantage. Most of real R&D falls into this category. So many applied 'AI' things are coming out of R&D divisions of big companies. 'Market Power' is a more abstract concept, but when you're large, you can make much stronger demands. Mondelez, P&G are big enough that they can plow through many competitors with all sorts of tactics: they get better prices and displays in-store, they have much bigger marketing budgets, they can sell at a loss in one corner to stifle a competitor, they can lean on buyers and distributors (i.e. buy them and you don't get our product) etc..You can think of Mondelez and P&G as 'owning' production and distribution. Apple is a 'Monopsony' for certain parts, and it has incredible leverage over it's supply chain and reaps most of the surplus profits from innovation in those supplying it. C) Many firms simply do not benefit from economies of scale i.e. roofers, carpenters, most kinds of law etc. but most do, however, the US has relatively speaking, much more people employed in 'big companies' that not - and that's what makes it rich. In fact this is a defining feature of 'Rich Europe' aka Germany, UK, Netherlands vs. 'Less Rich Europe' aka Poland, Spain: 'Rich Europe' has many more people in mid to larger companies. In many industries, you have to be at a certain scale to compete effectively. We generally don't like monopoly because the investors will tend to sit on an industry and extract rent as opposed to doing anything risky and actively block others from participating.

mudita 5y ago
I am not sure whom you mean, when you talk about "tech people”, but just in case you’re not aware: The author of the blog post is Robin Hanson, an associate professor of economics.

JetSetWilly 5y ago

Something like the NHS in the UK seems like a counterexample to this article's theory. A very large, extremely centralised state owned healthcare system, yet "health outcomes" seems perfectly in line with that of the intensely more expensive decentralised and privatised US system. And even in the recent pandemic, it was the NHS that discovered treatments like dexamethasone and tocilizumab thanks to an innovative and large scale A/B testing system that is much more difficult to set up in decentralised systems.

lottin 5y ago

The article talks about "concentration" not "centralisation". These are very different things. For example, a large organisation can be decentralised. This is not nitpicking, you can't think properly about stuff if you're using the wrong terminology.

xyzzy21 5y ago

Centralization also reduces redundancy which is essential to system reliability - a centralized system always ends up less reliable because there's more chance of single-point failure that cascades to all parts of the system quickly. This is especially true if "scale-free" features like "preferential attachment" exist. https://en.wikipedia.org/wiki/Preferential_attachment https://en.wikipedia.org/wiki/Preferential_attachment Consolidation of any type of thing (e.g. Wall Street and financial systems) can become very fragile and many of the guards against failure (risk models) cease to be valid the moment you have preferential attachment.

mark_l_watson 5y ago

I largely agree with the author, but they left out something important: arguably the most important thing that large corporations do to protect their interests is co-opting the government with campaign contributions, promises of future lucrative private sector employment, etc. Combined with control of the news media to divert attention, huge corporations have superpowers compared to small businesses and individuals.

zestyping 5y ago

This might be the sloppiest take I've ever seen on this topic. - "In fact firms do not do this." — In fact firms merge all the time. - "The typical scale of most firms seems far smaller than can be explained by these effects." — No reason is given. Italics do not constitute an argument. - "It seems obvious to me that by now such a food agency would have intervened in food production, processing, and distribution far more extensively that has been the actual case in our actual history." — Not at all obvious, and again no evidence or even any argument is given. - "The amazing thing is that all this would happen even with high quality oversight and accountability by agencies to politicians, and politicians to voters." — The amazing thing is that the author thinks he can state this prediction without basis of any sort. There are good reasons to beware centralization, but this piece is simply lazy. This gets an F. Show your work.