There’s a plau­si­ble case that cap­i­tal­ism per se isn’t the prob­lem. Con­cen­tra­tion is: Vari­a­tions range from “too big to fail” to good old-fashioned monopoly. It’s painful­ly ob­vi­ous that the ac­qui­si­tion of one mon­ster tele­com or bank or air­line by an­oth­er does nei­ther their cus­tomers nor the econ­o­my as a whole any good. It’s tough to write anti-monopoly law, be­cause how do you quan­ti­fy mar­ket pow­er? How­ev­er, you can quan­ti­fy size. So let’s use that as a club to bash away at the prob­lem.

Specif­i­cal­ly: Pick a num­ber X, and pass leg­is­la­tion de­cree­ing that no com­pa­ny can have more than X full-time-equivalent em­ploy­ees.

How big is X? · Let’s look at some data:

Em­ploy­ees (K)
Bri­tish Air­ways40
Gold­man Sachs30

The num­bers aren’t per­fec­t, most­ly sourced from Wikipedi­a, round­ed down to the near­est 10K above around 30K. Some peo­ple say Sam­sung has over 500K. The Tes­la num­ber is af­ter ac­quir­ing Grohmann and So­larCi­ty in late 2016.

I’m go­ing to say X is maybe around twen­ty thou­sand. Every sin­gle one of those com­pa­nies that’s 20K+ could, I think, be bro­ken up in­to san­er, sounder, small­er chunks that would be de­cen­t, prof­itable, high-quality busi­ness­es.

I guess maybe you couldn’t have a tele­com that owned a na­tion­al net­work and al­so had kiosks in ev­ery mal­l. Or an In­ter­net gi­ant that did email and op­er­at­ing sys­tems and cloud. Or a bank that op­er­at­ed branch­es in ev­ery town of a vast na­tion. Or an au­to­mo­tive com­pa­ny that sold a full line of per­son­al, recre­ation­al, and busi­ness ve­hi­cles on sev­en con­ti­nents. All these things seem OK to me.

The pic­ture would be weird­ly dif­fer­en­t: A suc­cess­ful young com­pa­ny ap­proach­ing size X would have to make painful choic­es about which busi­ness it re­al­ly want­ed to be in, and walk away from the rest. This doesn’t sound ter­ri­ble.

You could still grow fi­nan­cial­ly if you could fig­ure out a force mul­ti­pli­er for the abil­i­ties and en­er­gies of your em­ploy­ees. This doesn’t sound ter­ri­ble.

Com­pa­nies could still be ex­treme­ly pow­er­ful; look at Face­book. But I still think that, broad­ly, small­er com­pa­nies are less so. And fur­ther, it’s very like­ly that this mar­ket­place has a lot more com­pe­ti­tion than we see cur­rent­ly.

En­force­ment · We for­tu­nate cit­i­zens of the de­vel­oped world live un­der the rule of laws not men, but an un­for­tu­nate corol­lary is that busi­ness­es will game the rules right up to the edge and be­yond. So the rules would have to be pret­ty well cinched down.

  1. There’s no dis­tinc­tion be­tween em­ploy­ees and con­trac­tors; the num­ber of work­day hours you pay peo­ple to work can’t ex­ceed 8 times X.

  2. It’s not dis­tinct em­ploy­ees, it’s how many are work­ing on any giv­en day.

  3. (Anti-keiret­su):

    1. If any com­pa­ny owns more than 5% of the eq­ui­ty of an­oth­er, di­rect­ly or in­di­rect­ly, they are con­sid­ered a sin­gle com­pa­ny for the pur­pos­es of this leg­is­la­tion.

    2. If any com­pa­ny de­rives more than 25% of its rev­enue from an­oth­er com­pa­ny, like­wise.

    3. If any le­gal en­ti­ty owns more than 5% of the eq­ui­ty of more than one com­pa­ny, like­wise.

Pu­n­ish­ment · Laws are al­ways vi­o­lat­ed, and it’s crystal-clear that cer­tain sec­tors of the busi­ness com­mu­ni­ty see fines and oth­er forms of fi­nan­cial penal­ty as just an­oth­er cost of do­ing busi­ness.

So I pro­pose a sys­tem where there is on­ly one penal­ty. Should a cor­po­ra­tion be con­vict­ed of ex­ceed­ing size X, its chief ex­ec­u­tive and Board of Direc­tors at the time the vi­o­la­tion start­ed to oc­cur, and any suc­ces­sors, would be sub­ject to im­pris­on­ment un­til the il­le­gal con­di­tion is cor­rect­ed. So sim­ple!

This idea is com­plete­ly crazy and could nev­er hap­pen! · Over­ton Win­dow. First they laugh at you. Etc…


Comment feed for ongoing:Comments feed

From: Gavin B (Aug 31 2017, at 01:33)

Some sort of discussion going on over at the Guradian yesterday


From: Chris (Aug 31 2017, at 04:40)

Maybe, but I'm thinking that based on part evidence that companies that make physical products are less likely to be problematic. Maybe we need coefficients?


From: Ryan Baker (Aug 31 2017, at 05:42)

Interesting idea, but I'd be worried about the effects to a companies behavior around the time they approach the limit. Firing people probably isn't the best solution, but it would often be the quick fix employed. Jail time also seems a little harsh, so it will be difficult to enforce.

Alternatively, why not base this on revenues? In such a case there's two quick fixes that are more palatable, confiscation (by government) and price reduction (voluntary).

Besides which, the power of a company is mostly based on the amount of money it controls.


From: Raymond Lutz (Aug 31 2017, at 05:45)

I remember you're quite fluent in French so let me point you to this guy who brazenly submitted his candidacy for the présidence de l'Autorité des marchés financiers (en France) with those kind of "should work, won't pass crazy ideas" about finance regulation: I guess they didn't call him back...


From: John Cowan (Aug 31 2017, at 07:47)

What I think is bizarre and damaging is not just corporate size but corporate governance. We have found out that absolute monarchy, or even monarch + feudal council, is a crappy way to run countries. Why are we obsessed with it as a way to run companies?


From: Nathan (Aug 31 2017, at 10:37)

At a quick glance I count 8 different countries represented on that list. You'd have to convince all 8 countries to enact that legislation simultaneously or else companies in those countries that did enact it would just move operations to those countries that didn't. And it sounds just like a trick America would pull to totally pinky-promise they're going to enact the legislation and then conveniently lose it when other countries sign it off, thus causing other nations' biggest corporations to move over there.

Or perhaps all the countries on the list do sign it and then all those companies move to New Zealand so we can attain our rightful place in the world!

I note a certain large company conspicuously absent from your list. employs 340k individuals and they make an excellent case study for your theory. Under the iron fist of your regime they would likely be broken up into about 20 different companies, roughly half of which would likely fail within the first year. In fact, it's quite likely that Amazon Web Services would never exist, as rather than enjoying the advantage of being subsidized by Amazon's profitable businesses for the 7ish years that it was unprofitable on its own it would have had to exist as a startup and would not have enjoyed the partnership with Amazon's core business (not to mention the force of Amazon's brand and marketing) that it did in our universe.

An interesting thought experiment, certainly, and I sympathize with the underlying idea because it seems that megacorporations are poised to hold more power than most world nations, and here in New Zealand we see some of the effects of that as large American and Chinese companies with more financial strength than our whole country keep trying to exploit us. I'm not convinced that your idea as presented is a feasible way forward, but perhaps the combined might of the Internet cognoscenti will hammer it into a workable model.


From: Miguel (Aug 31 2017, at 13:29)

That seems like it would just encourage automation. Which might or might not be good. But it won't reduce the power of the giant companies.

Seems like you need some kind of cap based on the company's size relative to the country's GDP.


From: Robert de Forest (Aug 31 2017, at 15:28)

I work at Amazon, so I "should" be biased against a proposal to limit company size, but I agree with the motivating assertions about the dangers of large companies. I would even go father, but I'd be wandering into extremist territory and there's rarely much to be gained there.

I also agree with other commenters here that an arbitrary limit on concentrations of power, intelligence, etc, are impossible to enforce and have unwanted consequences when enforcement is attempted.

I propose to reverse the cause-and-effects driving these concentrations instead. I have no corporate finance experience, so my understanding of the dynamics is largely speculative. That said, it seems like the marketplace favors larger companies in several ways: economies of scale, tax reasons, resilience to surprises (lawsuits, disruptive tech, sudden market shifts, sudden political shifts), reduction of meaningful competition, lobbying strength, shared culture (everyone at Amazon who has been at the company more than a year references the Leadership Principles weekly if not daily) and the 900 pound gorilla effect: vendors tend to a customer who name is likely to come up in quarterly meetings than a customer they could stand to lose.

Some of the above reasons can also lead to benefits for customers, employees and vendors. Even if the company sits on the savings from their economies-of-scale, they're not putting that money in a matress. Eventually economies of scale benefit everyone because the represent a real reduction in waste.

Some of the above are already either vaguely or overtly illegal (tax dodges, monopolization, certain kinds of lobbying), but poorly enforced. I would support leaning harder on that lever to address those problems.

And some of the dynamics are much more complicated and/or nuanced.

I would be interested in seeing some ideas about how to re-balance our laws and behaviors to favor smaller groups working autonomously, but without losing the benefits of working as a single legal entity.


From: Rob (Aug 31 2017, at 20:16)

I was intrigued by your employee metric, but I guess it is to be expected of someone in the automation industry. The interesting thing is that the Venn diagram of biggest by: employees, market cap, and revenue, actually has relatively little overlap. Check it out, its a relatively straightforward google. (The much more interesting question of biggest by liquidity is of course somewhat harder to determine...)


From: Alcides Fonseca (Sep 01 2017, at 02:42)

Regarding the rule:

If any com­pa­ny de­rives more than 25% of its rev­enue from an­oth­er com­pa­ny, like­wise.

I believe this is unfeasible. Think about Mozilla. Their primary source of income is the default browser provider. More than 25% of its revenue comes from whichever engine they default to.

Additionally, there are many consultant companies (outsourcing) that have one large bank as a client that provides them with almost 50% of their income. Different owners implies different companies for me.


From: Ed Schechter (Sep 03 2017, at 22:47)

Interesting thesis It does leave open the question of where the money would come from to drive much of the next stage of technology.

Right now, all the technology bigs are throwing major resources at the coming machine learning epiphany, for better or worse. If they are broken up, and their partial monopolies cease to exist, and the resulting competition pushes prices to the no profit zone, it is not obvious who would be in a position to advance this work.

Government? There is certainly precedence in NASA and in the NIH's cancer research, but in the current climate I can't see the US government funding that sort of effort. Military? Maybe, but the results would likely not be published or usable (although the internet does largely owe its existence to military projects). Venture Capital? The risk would be large, the payout slow, and the prospect of having any resulting major successes quickly broken up would be demotivating.

So while too big is bad, we do ultimately benefit from some companies' abilities to set prices to the point where they make real money.

Now the challenge is to identify and favor these cases ...


From: Frank Wilhoit (Sep 16 2017, at 17:13)

If capitalism is the problem, it is because unaccountability is the problem.

Capitalism subject to the law is a contradiction.

What we have is mock-capitalism, in constant tension with the rule of law, a tension that plays out in a manner that is ad-hoc, improvisatory, purely wasteful, and discreditive to all participants.

To your other point, observation shows that Coase's ceiling of organizational size is somewhere between 60 and 100 persons. This applies to each *autonomous* unit within a larger, possibly hierarchal organization.


From: Mack (Sep 21 2017, at 01:00)

Couldn't you apply this principle to the original huge organizations, governments?

And what happens when you get coordination problems where the centralized organization is more efficient and outcompetes the decentralized organization?

If you look at the SF bay area today for example, you have a large amount of municipalities that all need to coordinate and play hot potato with each other. If the bay was one unified district, then when bart was constructed, it might of circled the bay originally. It becomes a tyranny of squabbling fiefdoms.

I feel like there is this continuous tension between centralization and decentralization, where centralization gives efficiency at the cost of freedom, but decentralization gives freedom at the cost of efficiency.


author · Dad · software · colophon · rights
picture of the day
August 30, 2017
· The World (112 fragments)
· · Politics (152 fragments)
· · · Radical Ideas

By .

I am an employee
of, but
the opinions expressed here
are my own, and no other party
necessarily agrees with them.

A full disclosure of my
professional interests is
on the author page.