Hacker News new | past | comments | ask | show | jobs | submit login
How Harvard Business School Has Reshaped American Capitalism (nytimes.com)
257 points by lxm on May 1, 2017 | hide | past | favorite | 210 comments



Whenever I hear critiques of MBAs I think of this anecdote from a paper [1] by Yanis Varoufakis:

In the parlance of signaling theory these [MBA] transcripts are known as screening devices. One personnel manager (of a financial services company) explained this to me candidly: “I have no doubt that economics courses, MBAs, etc. teach nothing I need my employees to know. If anything I need to reprogram them on the job. The courses they have taken encourage bad thinking.” “But then,” I asked, “why do you prefer to hire college graduates?” “Because,” he answered matter of factly, “if these kids were willing (and able) to pay large sums of money in fees; if they submitted boring assignment after boring assignment without fail, and did not complain vigorously in the process; then there is a good chance that they are the kind of people who will appeal to my customers and do as they are told.”

[1] https://www.yanisvaroufakis.eu/wp-content/uploads/2017/02/ag...


This is true of college degrees in general, lots of companies in various industries require proof of a bachelor's degree, regardless of what the subject matter was, because it shows an ability to stick to a goal and get something done, despite the actual quality or enjoyment of the work.


Spend tens of thousands of dollars, go into significant debt, just to prove you can "stick to a goal."

I can't be the only one who sees a problem with this. It's either a lie or a poor reflection on our society.


The problem with our society is that measures become targets. That's it, that's all there is to it. We keep failing to learn the lesson of Goodhart's law, over and over again.

It explains the politics of triangulation, bad monetary policy, the bad management style of MBAs, the emphasis on wealth for social signalling, and on and on and on. It's such a common phenomenon that I'm a bit scared!


The only solution is to raise good citizens that take pride in their work and strive to do the best through their own free will. It sounds like hippy BS, but we're never going to police everyone into doing a good job with metrics... might as well try to teach a different philosophy of work.


"Significant debt" being on average about the price of a low end car. And "prove you can stick to a goal" being a massive increase in lifetime earnings over those without a college degree.

I can't be the only person who sees how fair and progressive this is? It's a wonderful reflection on our society.


http://money.cnn.com/2016/10/18/pf/college/average-student-l...

Average debt for 2015 grads is 30k. That is not a low end car.

> And "prove you can stick to a goal" being a massive increase in lifetime earnings over those without a college degree.

Going to college is not the only way you can prove you stick to a goal, which according to the OP, is the only purpose of college. This is an immense waste of resources, all so that a company can take a little less risk when hiring someone, if at all. And college is such a business today that academic integrity is an afterthought, the degree is continually losing its value.

> I can't be the only person who sees how fair and progressive this is? It's a wonderful reflection on our society.

There is no need for sarcasm here. Either way, it doesn't follow from your argument that it is fair and progressive.


It took just a few moments to find out if $30K is actually fair and progressive. College grads average around $30K more per year in compensation than high school grads. These degrees are already highly subsidized. So tax payers, including those with only high school degrees, are paying for people to become educated and thus make more money, with the hope that some sort of trickle down economics will make it better for those without college degrees. I'd tongue in cheek call this highly regressive.


> It took just a few moments to find out if $30K is actually fair and progressive.

This is a false comparison. 30k is average DEBT they have-not the total cost of college education. Add many more tens of thousands-just so a company can take a little less risk, according to OP.

The reason we have subsidies is because businesses now want a degree, for no good reason, as evidenced by the OP. Because more people, who can't afford it, need college degrees, the government stepped in to provide that, with loans, which they make money on, and grants, which they hope to be paid for by future taxes. That doesn't account for all of the growth in tuition prices, rather, the demand does. Demand has skyrocketed because college grads make more. Since demand has skyrocketed, the supply of college grads has substantially increased-commoditizing the college degree, which benefits businesses because now they can pay less for a college grad. That's the bottom line.

The problem is more nuanced than just cost and salary. The fact is to make good salary college wasn't necessary decades ago. Now it is. So before what you got for free, now costs tens of thousands of dollars. How that is "progressive" is foreign to me.


Funny, I just realized my error and came here to correct it and saw you'd already done it. Hah, thanks. And I sympathize with the point about jobs paying more in the past without a degree, but I believe that the premise of the article is also wrong - that college degree confer no benefit to the company other than proving individuals can grind through boring tasks. Although to be fair, I got a CS/Math degree in 2.5 years, and I don't know that I've held a job where I couldn't have learned most of the programming required in a few months on the job. And I've never needed a lick of the math I so painfully learned. So maybe jobs actually requiring degrees are fewer and farer between than I realized.


Remember when turning 18 meant you were an adult? "prove you can stick to a goal" used to be what K-12 was for, before we decided not to fail kids because they weren't ready.

People should go to college because they are capable autonomous adults that require specialized knowledge for their career goals. It should be short, intense, and absolutely focused on that specialized knowledge.

If someone can't hack it, go back to grade school or seek some kind of professional help.


I don't know what world you live in where 30k is an insignificant amount of debt to put on a 22 year old


It's about a single year's difference in income between the average college grad and the average high school diploma. If someone told me I could invest something with a 100% return every year for the rest of my life, I'd likely not complain if there was a steep minimum buy in.

And I didn't claim it was insignificant only that it's what low end cars cost these days - which might be a bit off but not by much.


When you think about it, that's a horrible standard for hiring isn't it. Hiring someone who just follows the grain, does as they're told and doesn't complain despite the ridiculousness of the process. It seems to ensure that only the worst, least creative, most sheepish individuals will rise to the top. This could explain the slump modern society has found itself in.


It depends on why you are hiring them. There is always a place for people who will just take orders, and it doesn't mean they will rise to the top. If everyone was imaginative and didn't want to follow suit, it would be much more difficult to assemble enough people to cooperate and build something big together.


I disagree that this is a driving factor in getting college grads, even if it might be one of the reasons. College education gives you a wide breadth of knowledge and critical thinking skills that could be applied to many subjects. Its also just an accepted standard for many white collar jobs if one doesn't have the requisite experience.


This sentiment is further eloborated in the book "Disciplined Minds" - http://disciplinedminds.tripod.com/ I highly recommend it.


I also recommend it. It's especially interesting if you've been through a PhD or other graduate degree, since the author (who has a PhD in physics) spends a lot of time talking about how such educational programs indoctrinate their students.


Here's Chomsky expanding on this further: https://www.youtube.com/watch?v=pFf6_0T2ZoI


> WE EDUCATE LEADERS WHO MAKE A DIFFERENCE IN THE WORLD.

http://www.hbs.edu/about/Pages/default.aspx

> Careers change. Industries change. The world changes. With the programs at Stanford Graduate School of Business, you’ll not only keep pace with that change — you’ll drive it.

https://www.gsb.stanford.edu/stanford-gsb-experience/why-bus...

(Related: Cool URIs don't change https://www.w3.org/Provider/Style/URI)


Yanis Varoufakis is a hero.


The classical model of capitalism is that profits are a side effect of some socially beneficial activity. But today, "maximizing shareholder value" as a goal has been so well optimized that it often no longer requires doing anything useful. This led to such strange situations as 40% of US corporate profits coming from financial activities. (That ended in 2008.) There's still the hedge fund industry, which is a net lose for its customers but a huge win for its operators.

The Harvard Business School helped develop the theory, and the policy thinking, that made this possible. Short-term capital gains are not heavily taxed, for example. Both Bill Gates and Warren Buffett have said they should be. That policy, which favors financial maneuvering, is so well entrenched that it's not even discussed politically any more.


> Short-term capital gains are not heavily taxed, for example.

I think you mean long-term capital gains. Short term capital gains are taxed as ordinary income in the US.

[1] http://www.investopedia.com/articles/personal-finance/101515...


https://en.wikipedia.org/wiki/Carried_interest for private equity funds is taxed as long-term capital gains in the U.S. rather than as ordinary income.


Only for investments held in the fund for longer than 1 year. Read the Taxation section of your linked Wikipedia page again, it's petty clearly spelled out.


The classical model of capitalism is that profits are a side effect of some socially beneficial activity

No, the "invisible hand" as stated by Adam Smith, which is presumably what you mean, states the EXACT opposite causality order.


It's worth noting that Adam Smith was imagining an economy in which most market actors were small proprietors, a nontrivial minority were living self-sufficient lives, and very few were wage-laborers.

And then Enclosure happened -- precisely to create a landless "working class" who could not participate in the market as free actors!

https://newrepublic.com/article/141663/united-states-work

http://www.nakedcapitalism.com/2012/04/yasha-levine-recovere...


> It's worth noting that Adam Smith was imagining an economy in which most market actors were small proprietors

Is that a fair interpretation of Smith's work? The concept of economies of scale (dated back to Adam Smith) combined with laissez-faire approach advocated by Smith seems to imply a large corporate endeavor (Sears catalog, Wal-Mart, Costco, Amazon.com) would inevitably overpower any small-scale operation.


Could Smith have imagined huge multinational corporations in a time where economics was strictly mercantilist?


Just to chime in: the east india company was already a significant force by the mid-late 1700's, so yes, I do believe he had some image of the potential power of multinationals/megacorps.


I think even though 'Animats wrote this backwards, it doesn't really change the argument all that much. The way capitalism works may be "self-interested actions lead to socially beneficial outcomes", but as a society, we want to have capitalism precisely because of those socially beneficial outcomes. This "side effect" is the reason to employ capitalism in the first place.

Now it turns out (with-hindsight-20/20) that Adam Smith did not fully understand the "mind" behind the "invisible hand". It indeed does arrange for social benefits... except when it doesn't[0]. A good contemporary analysis of this mechanism is the article by Scott Alexander, "Meditations on Moloch".

http://slatestarcodex.com/2014/07/30/meditations-on-moloch/

--

[0] - For instance, a lot of those social benefits are in fact signs of market inefficiency, and those eventually get "corrected" as more and more people compete in a given sector and the margins dwindle.


[0] - For instance, a lot of those social benefits are in fact signs of market inefficiency, and those eventually get "corrected" as more and more people compete in a given sector and the margins dwindle.

Except for when you can get the government to step in and grant you rental rights preferable exclusive rights. Land, property, spectrum, licenses, copyright, patents, trademarks.


Not all interventions are good, but that doesn't mean there are no good interventions.


Not sure you can think of it as good or bad. The government takes away rights from someone people and grants them to others. I have the right to own land and make it so you can't use it because "we the people say so." good for me, bad for you.

I guess there's a net win now that I think about it. I could have protected with a club. good for me, good for you because I didn't club you in the head.


> I guess there's a net win now that I think about it. I could have protected with a club. good for me, good for you because I didn't club you in the head.

Yes, that's how it actually works. You have rights to some piece of land, I have rights to another, we don't step on each other's toes, and if we don't like the split, there are blood-free procedures to resolve disputes. More importantly, neither of us has to invest in clubs and constant vigilance, because we've outsourced that job. So we all save time, resources and effort.

It's not perfect, but it's a useful tool in our toolbox for building societies.


your land is protected by a club.

instead of protecting it yourself, you delegated the job to the government.

libertarians will argue "why not delegate to a private company?"

because we all learned from past experience. it is better for one entity to have a monopoly on violence than multiple small organizations.


What is meant by the invisible hand, and what did Smith say about it? I see this a lot, but I'm not sure what it means when I see it.

There are others with different views of capitalism, such as Marxian economists, Proudhon, neo-Ricardians, post-Keynesians etc. Marxian economists in particular with interesting things to say about the origin of profit, alienation, commodity fetishism and classes within capitalism.


"The invisible hand" is a notion that Adam Smith wrote about that essentially captures the idea of social benefits being an emergent consequence of self-interested parties engaging in the markets together.

You can read more about it here, including the (exactly three) specific passages in which Smith used the term: https://en.m.wikipedia.org/wiki/Invisible_hand#Other_uses_of...

The specific point of the phrase is that the individuals produce a net benefit to society through no intention of their own, it just happens to be a happy side effect of their work.

EDIT: Don't shoot the messenger, people.


How is this any different to "trickle down" on a large scale? If you encourage capitalistic self-interest enough, it will create benefit? Why not just create the benefit directly instead of going the roundabout way of having people act in their own self-interest, fostering mean-spirited relationships between people, competition instead of cooperation and general wastage of resources?

But perhaps there are other laws within capitalism. The law of value, for instance, which its recognition by capitalists and relative ignorance of it by wage labourers leads to Marx's theory of surplus value - this is the side effect of this self-interested activity of labour.


> If you encourage capitalistic self-interest enough, it will create benefit? Why not just create the benefit directly instead of going the roundabout way of having people act in their own self-interest, fostering mean-spirited relationships between people, competition instead of cooperation and general wastage of resources?

The idea is that this "roundabout" way is in fact more natural to individuals. People want to actively take care of themselves and their families first - you may say that "capitalistic self-interest" is actually rooted in our biology. Some forms of market economy seem to be the default, "lowest-energy-state" form of human self-organization. It has an added benefit of being heavily self-regulating through feedback loops implicit everywhere in trade.

Now, it would be wrong to pretend it all worked out perfectly and everyone is happy. It didn't and people aren't happy. Yet capitalism has the neat quality of aligning with a lot of basic human needs[0], and that's why while seemingly "roundabout", it's still[1] much easier than top-down management of social goodness.

--

[0] - Perceived autonomy would be another one of such needs.

[1] - I do wonder if in an interconnected reality of "information age" we couldn't pull off central management driven by algorithms. Someone on HN pointed out to me though that this would still lose an important feedback channel, namely preferences of customers as expressed by purchasing decisions.


> you may say that "capitalistic self-interest" is actually rooted in our biology.

I hold such deep issue with this. Sure, there's an aspect to all living beings that they "self-preserve." But you seem to be implying here that it's part of human nature to be primarily self-interested, and if that's a correct assumption, it's one I believe to be deeply incorrect.

We can look around and see people acting in self-interested ways all of the time, but we can also see people doing lots of benevolent, good, supportive things for other people. In addition, it's no surprise many people go around day-to-day acting this way, given that they live in a system which encourages (née requires) them to do so.

Just because people can conform and act within a system, does not mean it is part of their nature. A system that actively encouraged and promoted the common good and communal caring among individuals would likely produce more of that behavior as well. Just because we see one type of behavior or another is not enough to decide that it's human nature.


The way I see it: in general, people seem to form groups and care more about people in those groups than about those outside. What groups people form in their mind seems to be a pretty individual thing, depedent on the way one sees themselves in the world, but there are always groups and they always seem to be ordered.

That doesn't mean people don't give a shit about anything outside their closest clique! It only means that groups have different relative importance to an individual; for instance, when your family member is sick, you might temporarily stop attending the soup kitchen and spend that time helping the sick kin.

My point is that such "ordering of care" is part of our firmware, and because capitalism is compatible with that, it's easy to implement by humans - in the same way the basic concept of private property seems to be also a part of our firmware, and more extreme parts of communism being incompatible with that meant people started to work against the system.

Basically, if your system aligns well with natural tendencies of humans, it's easier to make it work. Corollary to that is that you should design your systems to work with natural tendencies of humans, instead of trying to change those tendencies directly.


Points taken. And I agree that we should design systems to work with the natural tendencies of humans. But we can do better than (and I'm being reductive here) capitalism's current "eventually everything will optimize out even if poor people starve to death because they weren't born with money." In fact, I'd argue that 'modern' capitalism really is only activating a small portion of human nature (that is, the self-interested part), and shielding us from those aspects of us that are caring for others and concerned for other human beings. When we put everything from the value of private jets to food, water, and healthcare on the same value metric (money), and then tell people that whatever amount of money they have is indicative of what they are worth, it's no wonder we as a society wind up treating people as objects instead of human beings.

Anecdotally, the US is a nation with more than enough food and shelter for all its inhabitants. Yet many people live on the streets, often because they were born into poverty. This goes against most people's "firmware." Yet capitalism takes no account of that part of human nature (I'd argue those parts that make us 'better than animals.').


I agree with your points here, and I'm not endorsing more capitalism as a solution to problems of current capitalism. I was only answering the question "why such roundabout way" by pointing out that, when viewed through the lens of human nature, it's actually a pretty direct way.

Though honestly, it's also easy to forget all the benefits capitalism brought us when criticizing it, and IMO it's worth to have them in mind when discussing alternatives.


I admit that self-interest is biological, I do not deny this; but I wish to posit that capitalistic self-interest is not biological, or at least, it is possible to overcome this. Capitalistic self-interest when applied to the capitalists thrives off the work of others, instead of working, he will use the work of others. In a similar way, the wage labourer, instead of working to secure what he wants, he can usually only work to secure what he needs in the form of his wage, which serves as his sustenance.

I have to also admit that capitalism is indeed very good to make sure that those who have needs have them met, provided that they are able to find the right person to sell their labour-power to, not too cheaply, and the rate of surplus isn't too high as to discourage the worker and alienate him even further from the things he makes.

Unfortunately though, I figure it's an unjust system because of the structural inequality of the trade itself; by far the biggest trade happening is that of the labour-power commodity for wage (starting off the M-C-M chain). But this trade is grounded within an unfair position of one over the other; the wage labourer needs that wage in order to survive, it is his means of sustenance, how he replenishes his ability to work (along with sleep and rest, which is also a desirable quality) and how he ensures future flow of people into the labour-power market by means of reproduction.

Although it would seem as though the worker "values" his wage more than his labour-time, the reality is that in many cases (pushed to the extreme in some developing nations) is that the worker values his life more than death; thus the only way to live is by selling labour (unable to by risk of death or lack of capital or social situation or having a family to feed to become bourgeois himself in most cases).

Now I'm not arguing that people don't need to work in a society free from wage labour - that wouldn't be much better than capitalism. It would require that the products of at least someone's labour are appropriated (which is what's happening now). Rather, the workers democratically decide what to make and in what quantity, and democratically decide what to do with the products afterwards, and what to do with any surplus product. With increased automation (used of course for workers to mean they work less rather than for more short-term profit). This sets up a voluntary system of exchanges, no one person oppressing another. Of course management, engineering etc. are also jobs that apply labour too.

With regard to [1], you may be interested in "Towards a New Socialism" by Paul W. Cockshott. He's written about this topic of cybernetic central management. He has other works too in the same vein. There are also alternatives such as decentralised planning and market socialism too, though I'm not really here to argue in favour of any particular one.

In another comment you say:

>My point is that such "ordering of care" is part of our firmware, and because capitalism is compatible with that, it's easy to implement by humans - in the same way the basic concept of private property seems to be also a part of our firmware, and more extreme parts of communism being incompatible with that meant people started to work against the system.

Capitalism (in the sense of the production of commodities) requires the division of labour, though the division of labour does not require capitalism. Besides, this, as Marx writes, humanity had existed for thousands of years with people doing the job of tailoring and weaving, yet there was no specific job for them. On the other hand, other primitive societies had division of labour but no production of commodities. Internal capitalist organisations also take advantage of the division of labour without commodity production (the sectors of a firm do not trade and barter or engage in M-C-M, despite the division of labour).

With regard to private property, I must disagree; I do not think it is natural at all. Personal property, the right of possession, may be considered a natural phenomenon, but to cordon off an area of land and engage wage labour upon it is not natural; the concept of property is a relatively recent one, having different but distinct forms under feudalism and the mester-slave society that preceeded that. Kropotkin and Proudhon have written about this in The Conquest of Bread and What is Property? respectively. Even if one will claim it is natural, Proudhon makes the case that it is founded in barbarism, and private property is incompatible with the ideals of freedom, equality and justice within a society. For this, he says "Property is impossible."

We should be cautious to decleare things as part of our firmware, especially given that studies on our 'firmware' take place in these modern capitalistic societies in which people are forced, so much that it is ingrained, to engage in wage labour unless lumpen or petit-bourgeois or bourgeois.

Apologies for the long comment, the moderators of Hacker News have put some kind of flag on my account which means that I'm not allowed to post too quickly, probably beacuse I get into heated discussion. I had to respond to your other point lest it happen again when I make a second post.


Trickle down referred to giving tax breaks to rich people so that they would buy expensive things (like Ferraris) and that this spending would spur economic movement because people have to sell, make, and repair Ferraris. The money they receive they would then use to buy groceries or a new TV. This is far different than the notion of the invisible hand, if only we counted government influence (deciding winners and losers).

> If you encourage capitalistic self-interest enough, it will create benefit?

Yes. Look around. I don't discount the role government has played when it does do good things, but the wonders of the world today are largely driven by the desire to do some work in order to gain some measurable economic benefit. Period.

You can't just create "this benefit directly". How would you do that?

I also think your general perspective on business is being warped by your obvious bias. Competition in industry does not necessarily lead toward mean-spirited relationships, or destructive competition. It can, but look at the tech industry, if anything, the opposite has happened.

Also, the wastage of resources is almost laughable. I say this not to be condescending, but because if there is one entity on the planet that is inefficient, it's government. Businesses go out of business if they don't use resources in the most efficient possible way. A government just prints more money, or if worse comes to worse, passes laws or regulations that distort the economy.

There are countries that do better when it comes to private-public cooperation, namely Scandinavian countries, but even those companies are capitalist countries and recognize that government does not have the capacity to guide economic development as a whole compared to the free exchange of goods and services. It's helpful once in awhile to jumpstart an industry, like solar/wind power in the United States, but government should largely stay out of business (aside from obvious regulation and protection).


Waste of resources is a good point. Both governments and private industries waste a lot, but the kinds of waste are different.

For starters, in such discussion I'd ignore money, because as good as it is for being a proxy for everything, it's still ultimately an arbitrary, abstract quantity. Printing money, in a way, is fine. We should talk about waste in terms of real resources - like energy and human time.

> Businesses go out of business if they don't use resources in the most efficient possible way.

Except often there are more businesses than strictly needed, and the surplus production is a waste. But more importantly, in a saturated market every bit of fuel, paper, metal, every kilowatt-hour and every man-hour spent on advertising is a waste of resources. Simple as that. The companies play zero-sum games on such markets, where every action is meant to cancels out effects of actions of other companies. Governments don't advertise, because they don't play the competition game (at least not internally).

Point being, there's lot of waste, public and private alike. Some is probably unavoidable thanks to thermodynamics alone, but it would be worth our time to consider how to build systems that waste less energy and human time just to operate.


Tangentially, I cannot help but add this: mining imaginary points using electricity is one of the most egregious wastes of resources the human mind has ever conceived of. Burning our planet because we cannot trust one another to issue currency equitably is wrong in the worst way.


Yes, that's my feeling too.

But in lesser way, this applies to a lot of our social and commercial systems - we're spending energy to enforce large-scale coordination between people. That spending could be entirely avoided if we could trust one another more. I doubt that in the nearby future we'll be able to make humans more cooperative by default, but I think it's worth stopping to consider just on what exactly we're spending energy, and maybe if there are ways to avoid that.

In particular, I shake my head when I see people excited about new "trustless" systems that replace inter-personal trust with cryptography. It's not a victory, it's a defeat. Maybe necessary, but still a loss.


High trust societies are typically ethnically or culturally homogeneous. It is far easier to feel empathy for someone that looks and/or acts like you.

I get the sense that this homogeneity is no longer seen as a benefit (for both good and bad reasons) and so expect that we are going to expend even more energy on coordination in the future.


I think improving equality or at least reducing poverty could go a long way to towards increasing societal trust, even in the presence of cultural differences.

Happier people are less angry, and thus less likely to hate the other, especially if the other has a similar quality of life to you. Not to mention economic background is a major aspect of cultural differences.

Also, pockets of cultural homogeneity would be less of a concern if people still had similar opportunities regardless of which one they fit into.


>Also, the wastage of resources is almost laughable. I say this not to be condescending, but because if there is one entity on the planet that is inefficient, it's government.

I'm not arguing for government, at least not necessarily. Although there are cases of the government forcing commodity producers to destroy their excess commodities, businesses by themselves also engage in these practices to maintain a favourable price with regard to supply and demand of their commodity.

>You can't just create "this benefit directly". How would you do that?

Workers working for themselves, receiving the product of their own labour, without engaging in the tedious exchange of commodities for money. Democratic decision as to how excess product is allocated and whom it is given to.

>government does not have the capacity to guide economic development as a whole compared to the free exchange of goods and services

I don't think we should be saying this as fact; it's useful to remember that central planning does not only come in the form of fixed five-year plans, nor is central planning necessary to rid ourselves of wage labour.

My "bias" may be that I am a Communist (which I find I am freuently downvoted for even mentioning) though I doubt any of us can say we are free of some bias. Having a position on a certain mode of organisation of society is by no means having a bias, in my opinion anyway.

https://www.wsj.com/articles/americas-dairy-farmers-dump-43-...


I'm an anarchist, which is usually a non-starter for many conversations, so I sympathize with being a Communist and having that distract from your message.

The reason I mentioned this bias was because the statements you were making appeared to be standard talking points of somebody who has already drawn a conclusion. I celebrate differences in opinion for political philosophy and didn't mean to imply that your stance meant anything bad.

> Workers working for themselves, receiving the product of their own labour, without engaging in the tedious exchange of commodities for money. Democratic decision as to how excess product is allocated and whom it is given to.

There is nothing to stop this right now. In fact, co-ops are becoming more regular, and bartering has been enabled by online systems. The US government allows people to buy products using Bitcoin, create worker-owned shops, and freely barter goods. Central planning means you can only do one of those three, and historically it has been done poorly.

>I don't think we should be saying this as fact; it's useful to remember that central planning does not only come in the form of fixed five-year plans, nor is central planning necessary to rid ourselves of wage labour.

I hate to state this as a matter of fact, but I think it is. Government cannot foresee the needs of the people beyond perhaps making sure everybody has a ration of energy or food or something like that. Central planning is just far too complicated and harsh. I think your only hope for a more communistic society is through technological innovation that drives the prices of goods so low that profit simply cannot be made.


>Why not just create the benefit directly instead of going the roundabout way of having people act in their own self-interest, fostering mean-spirited relationships between people, competition instead of cooperation and general wastage of resources?

Because people always act within their own self interest for the most part. One may take more than their fair share of the benefits of society while others will attempt to get out of paying theirs. Don't forget the officials whom will attempt to push those monies one direction or another in order to enrich themselves.

Capitalism is the only system that takes this into account as far as I can tell, everything else is either predicated on people willingly giving to help society on the large scale (counter to human nature) or forcing others to do so at the end of a gun (that tends to lead to dictatorships and suppression of freedom as it requires near total government)


>Because people always act within their own self interest for the most part.

That's not really true. People are, roughly speaking, reinforcement learners. This means that, yes, they respond to rewards by seeking further such rewards, but there are multiple kinds of reward. Financial profit is only one of those, and it actually has a declining marginal happiness once personal income goes over a certain level.

Other major rewards for real humans include being useful to other people, acting with personal autonomy, and exercising mastery (or learning towards mastery) in a domain of skill.

When we talk about economic value, what we really care about is the intersection: good work is useful to others (meets the customer's needs), and it lets the worker autonomously exercise mastery over a skillset (or otherwise, save their energy for domains in which they can do so), and it's profitable or self-sufficient with respect to resources. That's the psychologically real self-interest of the person doing the work, combined with the psychologically real self-interest of the person who needs the work done.

When you start disregarding working conditions and customer needs in the process of maximizing profit, your profit-maximization system has become useless for human purposes.


I was answering a specific question about the term, "invisible hand." You've raised valid points, but I wasn't trying to profess an ethical stance for relying on it one way or the other.


http://econlib.org/library/Topics/Details/invisiblehand.html

Every individual... neither intends to promote the public interest, nor knows how much he is promoting it... he intends only his own security; and by directing that industry in such a manner as its produce may be of the greatest value, he intends only his own gain, and he is in this, as in many other cases, led by an invisible hand to promote an end which was no part of his intention. Nor is it always the worse for the society that it was no part of it. By pursuing his own interest he frequently promotes that of the society more effectually than when he really intends to promote it.


keep in mind this was written prior to behavioral economics or Neuro economics or study of cognition.

understanding why people make the decisions they do and how they all fit in a large system remains a work in progress.

we certainly know more now to simply call it the invisible hand


Hmm that's interesting, in my view the capital view is the exact opposite. Namely, that people pursuing their own self-interest (in the form of profit) will generate socially beneficial activities as a side effect (because society is in the long run not willing to pay for harmful activities).

I agree though that the financial sector has grown more influential than it should be and that society as a whole seems to greatly prefer short term than long term gains.


If this view is correct then you need regulation as some areas of the market are zero sum (or worse negative sum) and self-interested gain in these areas necessarily means losses to others; and for unmeasured externalities this can lead to very very large public losses, pollution is one such example.

In my opinion the vast majority of people prefer to "grow the pie" and win-win situations; but again equality of outcomes also figures in individual / social perception.


A good question here is "What is the best form of regulation?" In some cases, media focused on business activity or industry groups perform better at regulating the market than government. One example is the food industry. An outbreak of salmonella or other food borne illness that is heavily reported by the press is far more damaging and a much stronger inventivizer for food manufacturing facilities to maintain safe production standards than the FDA or USDA.


Media is in reaction to multiple people catching deadly illnesses. Media can also be silenced. Worse, it's not even guaranteed the media catches on that. And it'll be extremely localised.

FDA rules prevent those events from happening, as well as guaranteeing a baseline for food quality.

I'm going to stay with the FDA.


> FDA rules prevent those events from happening

If that was true then we wouldn't hear about it in the media. We do hear about it, which means that it does still happen, which means that the FDA rules do not prevent those events from happening.


Murder is illegal as well, yet you still hear about murders. That doesn't mean that the police aren't effective. Public health laws prohibit restaurants from storing raw meat at room temperature in the kitchen. Inspectors routinely find violations, and remediate by fining or even shutting down the restaurant.

Likewise, the FDA and USDA regulations are both largely effective and provide a common compliance target for producers and manufacturers to achieve. Sometimes there are issues, and the government has the ability to fine or demand recalls to protect public safety.

Many of the systemic gaps, especially in meat production, are a result of de-regulation where the USDA/FDA lost the ability to recall tainted meats and can only request voluntary compliance.


You're right; we do still hear about murders. We hear about them because the law does not prevent them from happening.

I was not trying to claim that the FDA never does any good. I was directly countering a false claim presented by the person to whom I was replying.


Since you want to be pedantic, yes, it does not _prevent_.

It helps with it not being a daily occurence. Which it would be, were the FDA not there, or had there been no regulations.


You don't hear about all the times that the FDA actually stopped tainted food products from reaching the market.


The media get the official notification from the FDA or USDA. There is a scientific process being followed from an authority that's accountable to the public regardless of profit motive.

There are such things as market failures, and at least in a democracy, it's a fairly quick event that market failures lead the public to petitioning the government to legislate alternative outcomes to the market.


And yet good food regulations have saved many more lives from outbreaks than all the negative media combined. The media is good but it works best on saliency, the unusual. People, a very very large number, die every day in car accidents, but it rarely hits the headlines unless it is an unusual case.


I'd like to see some statistics to back up this assertion.


I must admit I don't have any comprehensive stats to hand, even the WHO have only recently compiled worldwide occurrence stats but not correlation / causation stats.

However from a programming paradigm, which would you prefer? Bugs cleaned up only because "someone" reported them, or because you use CI and can fix the problem at source, can measure testing coverage, etc, etc. I think most programmers agree that the proactive systemic testing system is better than ad hoc.


So is that proactive systemic testing put in place because the FDA requires it, or because the food processing facility fears an outbreak that would bankrupt the company?

There is no FDA for bugs in software companies, which is far less consequential, yet most software companies invest a fairly reasonable amount of engineering time into quality control. So this begs the questions, how much good does the FDA really do?


A free market is not a market without regulations, in the same way as a free society is not a society without laws.


This is a specious analogy. A free society is free because it can decide its own laws. The equivalent would be self-regulation by market participants.


Considering that the whole country are market participants and not just the capitalists they can create market rules the same way they can create laws and still be free. The whole belief that the "markets" are apart from society makes no sense.


Just like a fox is self-regulating at the hen house?


And this happens because there is a dichotomy; a free society with only one class of participant vs a society with free people and slaves. The marketplace always has at least two classes, buyers and sellers. Buyers are typically those with liquid negotiable assets and sellers those with less liquid / negotiable assets; the most common liquid negotiable asset being cash. So when there is more than one class then there is a much higher probability that one party will have more "power" than the other, cf voters vs representatives in representative democracies, multi-national companies vs individual consumers.


The problem with your argument, that 'society is in the long run not willing to pay for harmful activities' will cause self regulation in an untrammelled capital market, is that the information and execution systems just do not exist for society to make such decisions in an informed and timely manner.


>> Short-term capital gains are not heavily taxed, for example.

This is not correct, at least in the USA:

https://en.wikipedia.org/wiki/Capital_gains_tax_in_the_Unite...

"Short-term capital gains are taxed at the investor's ordinary income tax rate and are defined as investments held for a year or less before being sold. Long-term capital gains, on dispositions of assets held for more than one year, are taxed at a lower rate."


"The classical model of capitalism is that profits are a side effect of some socially beneficial activity."

What you are describing is not really capitalism at all.


It is, though. The classical capitalist system is one in which people engage in commerce voluntarily. If both parties do not benefit from a trade then the trade does not happen.


two counter-parties benefiting from a trade is not the same as "society" benefiting from a trade


> Both Bill Gates and Warren Buffett have said they should be.

So? How does this appeal to authority meaningfully prove that it should be done? I'm sure Bill Gates and Warren Buffett believe a lot of things. Two (extremely) rich guys saying something is not a proper foundation for economic policy.

You didn't actually stake a claim about capital gains and its causal effect on quality of life in the United States. Can you point to something concrete, instead of just dropping in words like "hedge funds" and "2008"?

For example, something you might start with is the extent to which hedge funds have been a net loss specifically for pension plans, retirement IRAs and university endowments across the industry (that's off the top of my head, I don't know how effective an argument this is). This is one direction that I could see the financial services industry getting overzealous, in that the pursuit of fewer investors to manage with greater pools of money leads to institutional aggregations of non-accredited investors being impacted. But this strikes me as an issue with that issue specifically, not capital gains.


It's reasonable to be suspicious of anything that suggests taxation as a prescriptive treatment. Government force is like hard-coding a value in the code; if you really must do it, it may be OK, and everything needs something roughly equivalent in some form to provide the basics (variables loaded at runtime from the environment or a config file, etc), but it's never really the answer you want to hear.

I agree that our specific investment system incentivizes eat-the-world behaviors that are ultimately socially destructive. I am pro-competition, but just as we don't insist track and field athletes bludgeon their slower opponents to death, I don't think there is an economic imperative to heavily favor an investment/banking/financial class, especially when they provide no tangible or real economic benefit. Their work is all in pushing paper and moving numbers. In a well-aligned system, there is no way they should be capturing the amount of value they capture (and same goes for copyright interests).


I'd argue that this class _does_ provide tangible economic benefits. Unfortunately, there are some folks at the margin who don't. The difficulty is in drawing that line.

Apart from the liquidity, the large algorithmic firms have traditionally spent like crazy on computing. It was this hardware spend that allowed Sun to pursue a strategy where most of their software was open source.

Has Wall Street harmed you more than having freely available Java, Solaris forks, and ZFS have helped?


Or when you do discuss/question it, people try to paint you as a quack.


The modern MBA is so efficient that is has optimized out research and development and many times the building part, which leads back to r&d. This is great for short-term and the upper end, bad long-term and for the core product. A big part of the equation missing today is longer term consumer strength i.e. wages, community impact, continue self-disruption/research etc.

Startups usually do most of the r&d now, and once it gets to small/mid level and "the process" comes in, the bizdevs take over after it is usually already an established product and r&d mostly completed. Depending on the balance of that takeover it either kills the product, slowly bleeds it, or if the focus is still on product then sometimes a super growth engine.

Companies that can keep a focus on r&d, building and changing markets while being very efficient are key. Usually these are engineer led (Amazon, Tesla, Google etc)


Google is an example of the failure of being engineer led in my opinion. I think it's led to Google having institutional ADD and has left them with countless failed products and others that have been neglected for years (has anyone worked on Google Finance in the past 5 years? Probably not).

A lot of successful Google products were acquired (Android, YouTube) and others were able to leverage their dominance in search. But we also saw that approach fail when the entire company was directed at trying to beat Facebook. It infiltrated all their existing properties and employees were directly incentivized with compensation linked to the success of Google+. And yet all of that wasn't able to overcome Facebook's network effect. The go-to Google strategy for creating a hit product failed (and it also explains why they seem to fail at anything that has network effects or requires consumer affinity and brand: 8 messengers and countless failed hardware products).

Google stands a good chance of being the next Xerox PARC: good at science projects and technology, but destined to see others capitalize on it.

And it makes sense if you think about it. Google's origin story was 2 PhDs coming up with better technology in a market that was already defined and a business model that was already understood. They applied that technology well, but it was only possible because of the openness of the web. It feels very much right place, right time.


My view is that Google is insanely good at products that demand performance at scale. Google search, YouTube, Google Docs, Google Maps, Gmail, and anything that supplements their usage Android, Chrome. Their failures like Google+, Google Glass indicate a product-market misfit than a result of being engineer led. An engineer (software related) is no less fit to lead a tech firm than a doctor is fit to lead a Medical service facility than a musician leading a record company than a sportsman managing a sport team than a chef leading a restaurant.

Sure it could be the next Xerox PARC, but the next Apple (following the bad metaphor) will very likely be a division of Google itself.

And finally, if anything, Google's ability to kill products that it does not think valuable, is what keeps it innovating, exploring and yet not be a dinosaur search company. (I do miss Google Reader though. Did they really have to kill it so bad ? )


Isn't Google amorphous by design? I thought Google was supposed to exist everywhere hidden in plain sight, and its many products are just the interfaces it uses to connect to us. At least that's the impression I got from listening to Schmidt all these years talk about how Google can go anywhere and be anything it wants to be.


The counterexample is Orkut, Google's homegrown, initially successful social network, years before the rise of Facebook. It died mostly from scaling issues, i.e. became slow when many users were on it.


If Google is an example of "failure", I'm pretty sure a lot of other companies would love to fail.

> Android, YouTube

First off, Android didn't even exist as a product yet when Google acquired it, and YouTube was popular but nowhere near the behemoth it is now. Taking those things and turning them into world beating juggernauts is a very non-trivial task; as you've probably noticed, most corporate acquisitions of startups have a very different outcome.


Google had search and used that to prop the rest. Simple.

And Google is making money by removing more and more "free" listings and replacing them ads and their content. Morality aside, eventually they'll lose their effectiveness.


> Google had search and used that to prop the rest. Simple.

If it's so simple why aren't there other companies doing the same?

> Morality aside, eventually they'll lose their effectiveness.

People have said this for the last ten years. Sure, if you cry wolf long enough you will probably be correct at some point, but does this offer any insight?


I can't imagine anyone who's used the YouTube Android app would think it is skill rather than a massive existing userbase & network effect that's led to YouTube's success.

Google is absolutely horrible at UI and application design, I can't think of any product that doesn't mostly suck from a UI perspective. Scale & speed is obviously a different story.


What do you think about the UI of their flagship search website, goggle.com?


It's rubbish!

When I load Google.com, I should see the Google logo and at the bottom of a page a "click to continue reading" button that I should, well, click. Then as I continue to scroll down the search entry field should scroll up the page as the Google logo slowly fades out behind it. Searching should feel like an event.

But seriously, I love that Google keeps their search website so minimal. I think that takes courage.


It's fast, but it has many shortcomings.


He didn't say what they did was simple. He said it's a simple explanation for what they did. Google created a great search engine at the right place and time. Since then Google's products have been a hit but mostly a miss by a company with endless money.


Having many misses is just a result of trying many times. Hell, we're in the forum of the epitome of the idea that having many misses and a few hits can be a tremendously successful strategy.


The line should always be moving. Either up or down, hopefully mostly up, but moving nonetheless. It is perfectly fine for the line to go down, failing is learning. But if the line continues to go down, you're not learning, you're just failing. Fix that. Same as when the line continues to go up. If it goes up with no dips down, you're not learning either. Fix that.

The line should never be level. This is the worst position to be in. It means you're not doing anything at all.


>> Google had search and used that to prop the rest. Simple. > If it's so simple why aren't there other companies doing the same?

Because the "simple" strategy is predicated on a market-dominant search product. It was simple for them, but extremely hard for anyone else—because Step 1 is "supplant Google with 75% market share".


As it turns out the engineers are good at business.

Using your established brand to propel a new product is business 101 and building a whole web of closely related products is business 302.

The synergy between their many major products is quite astonishing (though it is not public knowledge).


Nevermind...


> Google had search and used that to prop the rest. Simple.

Simple? Tell that to anyone else doing search.


Talking about Google products one has to mention the two products which make around 90 percent of their revenue: adwords and adsense. These programs are the real products of Google, and at least one of them was acquired. It must be really difficult to do other products in an organisation that makes more than 50B on those two products alone. No matter what product or service you launch it's bound to make peanuts compared to these products for 5-10 years to come. Worldwide there is barely a handful of products that come close to adwords and adsense in revenue. How do you get the attention and ressources from management when you are up against that?


One of the most successful companies in history is an example of failure in leadership to you? Thanks for the lawls


> I think it's led to Google having institutional ADD and has left them with countless failed products and others that have been neglected for years

You're missing the point. I think this is part of what defines their existence and probably their success; experiment with many different products/services (and do it quickly too), doesn't matter if you have 1 killer product for every 10,000 experiments.


> Google is an example of the failure of being engineer led in my opinion.

For me, Google is a failure under the very same rule that makes MBA-driven businesses a failure for society. I.e. I'd actually prefer more business to be Google-like "failures".

For example, talk what you like about how Android is crap, but who'd you trust more with mobile OS - Google, or Samsung? Siemens? Oracle?


How are failed and neglected projects a result of being engineering led? If anything, I would think that killing and deprioritizing products, as well as making revenue fueled decisions (i.e. google shopping express for google home) are examples of management making potentially harmful decisions.

I will agree that Google seems lackluster at network effect related things and I find that they are generally very poor at marketing.


> has left them with countless failed products and others that have been neglected for years

Building things that you consider failed is going to be a natural consequence of finding things that work because you gotta try things to know. As for Google finance if it isn't broke why change it? Don't confuse change with improvement.


No MBA was EVER designed for the R&D part of building a company. I'm constantly surprised at the myth that an MBA has anything to do with teaching you to start or even run a company. MBA's are designed to train middle managers because the modern "firm" realized great middle managers were really valuable (aka they'd pay them a lot) but hard and risky to train. That's where the MBA program comes from.

MBA's are ANTITHETICAL to entrepreneurship. That doesn't mean an MBA doesn't have value towards starting a company, just like a degree in economics or history aren't irrelevant to starting a company. But it's not what it's for, in fact, it's for the opposite thing.


This is a gross generalisation (by which I mean, I think you are wrong). My MBA set me up perfectly to quit my job and launch a startup. Maybe yours was designed to turn out middle managers. Mine wasn't. I had ten elective subjects and they were all towards starting and growing a service business. And the core subjects have proved an invaluable primer in many matters of strategy and day-to-day operations, ones in which I would otherwise be winging it.


Launching a startup does not equal R&D. They have things in common but they aren't the same. And the only convenient thing, I believe, a full time MBA provides is a starting point to build your network.


On average MBA's are less likely to start a business than someone from the general population, you are the exception.


From FT: Overall, 19 per cent of MBA graduates from the class of 2012 started their own company.

This is the average over all schools in their global ranking, and the US has most positions in the top 25 [1].

Meanwhile, only ~300 persons for every 100k start companies per month in the US [2]. Even assuming none start more than one per four years (which is doubtful), that's only 14.4% of the overall population over four years.

[1] https://www.ft.com/content/cbd9b418-314c-11e6-bda0-04585c31b...

[2] http://www.kauffman.org/~/media/kauffman_org/research%20repo...


Than an average engineer? Maybe. An average someone from general population? Citation needed.


This is irrelevant to my point, just to be clear. Whether they're effective at creating founders is debatable, whether they're designed to create founders is not.


You made a false and sweeping statement, and now you're doubling down on it with ever more preposterous absolutes. Your understanding is evidently limited to a grossly distorted and narrow perspective on the MBA curriculum.

In reality, business schools around the world turn out people who are qualified to start and run a company, and this includes R&D processes. Some people may choose electives that lead to middle-management, but others can and do choose electives in starting up, product development, service process design, negotiations etc etc.


I think you're continuing to misunderstand my statement. You're suggesting some MBA programs are good at producing founders. I have no idea if that's true and no opinion. The MBA and all traditional MBA programs were started to produce managers at large companies, there are no exceptions and it is well documented and easily researched.


You said, and I quote: "No MBA was EVER designed for the R&D part of building a company".

This is garbage. Unmitigated, sweeping, unknowing drivel. I understand your claim perfectly well, and you are flat wrong, and apparently from a position of zero awareness. I can correct you authoritatively because a) I have actually completed an MBA, and b) it was one that included comparative studies of R&D processes in the case work and a team-based competitive simulation of managing R&D processes. Along with a toolkit of knowledge that made company formation and launch much easier. This has, as I say, all been super useful in actually building my business.

Moving the goalposts to some nineteenth-century context and claiming it's the same thing is, as I said, simply doubling down on the wrong. Also irrelevant.


Your MBA program (or some other MBA program you know of) was designed to specifically to give people the skills to found companies? I doubt strongly that it was.

The only exception to this I know is Babson, but I'm not certain whether they qualify their entrepreneurship degree as an MBA or not. I think they don't, but if they do I may have missed the mark by 1. Otherwise what I said is correct.


See, there you go again. Changing your statement once more - this time to something that is so specific and twisted from reality to be meaningless, and pretending it is the same thing.

I think I'm done here; having presented real-world counterexamples, your claim, which was a gross and simplistic misrepresentation, is dead.


You think I'm moving the ball because you're convinced I'm arguing something I'm not arguing. You're convinced I'm arguing about the value of an MBA rather than why they exist and what they're designed to do. It's possible other people upvoting my original post also misunderstood my point.

You think I'm moving the ball to something meaningless because you want to debate the value of an MBA and I don't.


Just out of curiosity, where did you receive your MBA & what electives did you take that you found the most value from in regards to starting and growing your service business?


Melbourne Business School, class of 2015 (part-time).

The electives that have been most useful to this startup: "Entrepreneurship & New Value Creation", "Managing Service Businesses", "Negotiations", "Mergers & Acquisitions", "Business Law".

Amongst other outcomes, these units a) quite radically changed my perspective on matters pertaining to value creation & capture - which was especially surprising for someone with two decades meandering experience across many different product & service development gigs - and b) gave me the analytical tools & confidence I needed to put my ideas into practice.


It's not a generalisation it's a fact. The fact that it set you up well to launch a startup is a fortunate coincidence, period. MBAs not being designed to help people start companies is a matter of historical record it's not up for debate to be honest.


It's said of the Harvard Business School that it trains people for two jobs: President and Chairman of the Board.


..which demonstrated that it's primarily about the network, and not about what you're learning


If that's all true then why did Harvard open a school to train middle managers?


Efficient? You’re too generous. I’d call them short-sighted, or even go as far as to call them looters: they are people who gut companies (again, the word “optimize” is too generous!) to boost short-term profits and line their pockets at the expensive of long-term viability in ways you mentioned (cutting out R&D, building of product, etc.)


I call them robots.

The robots are great at doing one thing well by hook or crook. Problem is with scale and speed that come with large orgs, the good little robots follow their programming and over optimize the heck out of one little variable. The environment has no time to adapt. Massive unrecoverable collateral damage happens. But the robots are all very pleased with themselves cause their one little self worth defining metric has been met. But as nature already demonstrates highly virulent viruses like these cannot sustain themselves long term. This robot class will die out sooner or later cause the cost for destabilizing the ecosystem every couple days is just too high.


Nope. That robot will find another place to go and optimize another variable.


Virulent viruses don't work like that. Once they destabilize one subsystem, the system as a whole takes hit. By the time they switch to which ever other subsystem they want the body is more or less shutting down. That's why highly optimized virus don't last long.


Maybe we are lucky that admission rates into MBA programs are not that high.


The MBA thinks in spreadsheets and talks in scalpels.


Happily, about 90% of business is spreadsheets and scalpels.


Heh. The fish does not notice the water.


I have a Harvard MBA ('14, sec C), and where this whole argument falls apart for me is the idea of "The Golden Ticket". The author is confusing correlation with causation.

My MBA has had virtually no effect on my opportunities or employability. Heck, as an engineer and entrepreneur, it has probably done more harm than good. This is a consistent experience - with the exception of (perhaps) the military and overseas students folks tend to go back to their previous career track and trajectory.

Furthermore it's not really a very transformative experience. The ideals, priorities and abilities students bring to the program are generally the ones they leave with.

What HBS does well, above all else, is act as a filter. They really do a remarkable job at admissions, putting together a group of folks who are likely (by nature or nurture) to go on to have an outsized impact on the world.


> My MBA has had virtually no effect on my opportunities or employability... What HBD does well, above all else, is act as a filter."

And you don't think employers realize this? You don't think this pre-screening process impacts their judgement when looking at job candidates?

The top management consultancies, the top investment banks and private equity groups are full of HBS grads. You think they all would have got these jobs without going through the HBS process, networking with their peers and alumni who end up working for these same companies, interning, and getting the HBS seal of approval?

You refer to yourself as an engineer. That might be why HBS didn't help your employability. But if you don't think it helped others, you're dreaming.


> The top management consultancies, the top investment banks and private equity groups are full of HBS grads. You think they all would have got these jobs without going through the HBS process, networking with their peers and alumni who end up working for these same companies, interning, and getting the HBS seal of approval?

Well yeah, absolutely. HBS is full of analysts and consultants from those firms and institutions taking a pit stop on their way to partner.

I find it fascinating how many people here are eager to tell me I don't know what I'm talking about when I'm the one with firsthand experience..


Let's take a specific example. For grad school, I went to a university with a generally well regarded MBA program, which drew on-campus recruiting from MBB. The recruiting process for MBA students was short recruiter interview -> case interviews. Non-MBA students interested in positions at MBB had it a bit different: in between the recruiter interview and the case interviews was an extremely difficult written exam designed to test the candidate's "reasoning" ability. A couple of my (pretty smart) classmates tried to prepare for these exams and eventually took them, failing miserably. They both reported that it was the hardest test they ever took. Both ended up at second-tier consulting firms.

Let's get more general and rigorous. Kellogg Business School professor Lauren Rivera studies the effects of educational credentials on employment opportunity. Her research confirms that educational pedigree is the single most important factor for gaining employment in elite positions in finance, consulting, and law, with the implication that your education is an indication of your social class, and your comfort in the upper echelons of society. [Link to her work] (http://www.kellogg.northwestern.edu/faculty/directory/rivera...)

I would say it's bit ignorant to say that a Harvard MBA has no effect on employability or opportunity.


Comparing MBAs' & undergrads' experiences is specious, at best. And if you're comparing other graduate students against MBAs, the recruiting process for MBB is exactly the same. (source: I did my time at McKinsey, the 'M' in MBB.)

Regarding Prof. Rivera's study, I'm curious whether it attempts to separate causation & correlation. At HBS, the number of ex-MBB consultants in the program is roughly equal to the number who leave into MBB - many swap out and in, but the figures are in the same range. Does she account for that?

Also, any analysis which groups JDs & MBAs together is 100% flawed. Those two may have similar prestige, but wildly different career progressions.


> Comparing MBAs' & undergrads' experiences is specious, at best. And if you're comparing other graduate students against MBAs, the recruiting process for MBB is exactly the same. (source: I did my time at McKinsey, the 'M' in MBB.)

I'm talking about other graduate students, and (at least in this example) the recruiting process is NOT the same. It didn't matter if you were getting a master's in basket weaving or a PhD in physics, you were required to take the test, again at least at this particular university. MBAs were spared the test and went straight to the normal interview process. Also, had you not done an elite MBA, what do you think your path to McKinsey would have consisted of?

> Regarding Prof. Rivera's study, I'm curious whether it attempts to separate causation & correlation. At HBS, the number of ex-MBB consultants in the program is roughly equal to the number who leave into MBB - many swap out and in, but the figures are in the same range. Does she account for that?

Prof. Rivera's research consists of multiple related studies whose findings confirm that, regardless of skills and relevant experience, the most important factor in hiring for elite firms in business and law are educational pedigree, creating a discrepancy in opportunity for those that get an elite education and those that don't. She accomplishes this through ethnographic research on the recruitment practices of the firms.

> Also, any analysis which groups JDs & MBAs together is 100% flawed. Those two may have similar prestige, but wildly different career progressions.

I disagree in this context. The point of her research is that careers which tend to lead to powerful positions in society are practically closed off to those that don't have elite educations. If you want get into high finance or MBB consulting or BigLaw, it behooves you to attend an elite university, and while you're there, do as the Romans do. Prof. Rivera's research is _really_ interesting, you should check it out.


> HBS is full of analysts and consultants from those firms and institutions taking a pit stop on their way to partner.

Many get their MBA to help them move up from analyst to consultant. Most won't make partner, but many will spin into lucrative executive positions with Fortune 1000 companies. So yes, some already are on the inside track with top firms, and need HBS or another Ivy MBA to move up to the next level. For others, it will help them build credibility and get their foot in the door. In either case, HBS is acting as gatekeeper.

When you look at how much the banking and consulting industries focus their recruiting efforts on a handful of top schools, I don't think correlation or causation is even a question. Graduates from other schools are not getting the same attention (perhaps because HBS has done such a great job screening their applicants already), which further helps HBS grads.

I'm sorry HBS didn't help your career as an engineer/entrepreneur, but this doesn't negate the experience of the 50+% of students who go into banking and consulting positions, for whom an HBS MBA and the alumni connections will be a boon.


Thank you for sharing your story and putting up with these comments.

I just re-visited this thread a day later and didn't realize many of your points were not only shared by my comments in parallel, but also being re-requested by the same people.


>You think they all would have got these jobs without going through the HBS process...?

Despite only referring to his experience, I think the extrapolation holds for most HBS grads( i.e. If they have the skills and early success trajectory to get into HBS, they could get similar offers and opportunities without an HBS stamp. Don't discount the two years' opportunity cost.)


Is this the case? I know, for example, in law, your employability at desirable firms is directly linked to an aristocratic educational pedigree. Is there a reason this would not be in the case in business? While I'm sure potential HBS attendees could find something more valuable to do with their lives in those 2 years, would they really end up in investment banks and private equity firms?

These seem like employers who are all about credentials, since there's no real way to demonstrate a propensity for gutting companies or engineering IPOs a priori, and since these groups are run on autopilot by corporate drones instead of people with a real sense of ownership or entrepreneurship (and thus, a desire to acquire and retain good talent regardless of its pedigree). In such cases, the hiring seems to become 99% about the value placed on the institution that granted the degree, and 1% about showing up in an adequately expensive suit and car.


Prestige matters in finance, too. However, a JD's not really comparable to an MBA because (a) getting the post-MBA jobs you mentioned typically require prior, relevant, gold-plated work experience (Switching into these finance sub-fields is VERY hard even for HBS MBAs.) and (b) these jobs don't require an MBA (unlike many jurisdictions that require accredited JDs). JDs typically don't have gold-plated work experience and have to rely on their law school's brand for signalling.

In other words, most HBS folks who choose this high finance route already have the credentials and pedigree for the job without needing an MBA.

For the rest, I recommend googling HBS job reports and re-thinking a few of your assumptions about where/how people create value at firms.


> they could get similar offers and opportunities without an HBS stamp

Many HBS students work as analysts at management consultancies prior to their MBA, and their employers send them to HBS before promoting them to consultant. What happens when the analyst refuses to get an MBA?

The banking sector also has a recruitment funnel at the top MBA schools. While I'm sure it's possible to get your foot in the door without an Ivy League pedigree, many recruiters are focusing their attention on a handful of schools.


For consultants, think about the mechanics of getting accepted to HBS. You need a partner to give you an LOR. That speaks well for your ability to get promoted. These analysts rise without an MBA or leave for other ops.

I think you're referring to a subset of FinServ (i.e. banking) with homogeneous roles aimed at MBAs. Again, if you get into HBS, you're qualified/networked without needing the MBA. The banking cohort isn't as pedigree focused as other niches within FinServ.


Maybe they would rise. In that case, why do they feel the need to do it?


People who go on to have an outsized impact on the world don't usually participate in programs that are neither useful to their careers (to the point of being detrimental) nor transformative to their lives.

If a Harvard MBA is essentially pointless like you describe - people go into it and come out the same - the only reason to pursue it is to buy into the network.

Hence, perhaps, the "Golden Passport" nomenclature. You'd also notice the author didn't write "Golden Ticket", which I dare say, is an interesting freudian slip.


It's a good two-year summer camp at a time in our career when we're all racing towards our first burnout, and a good opportunity to hang out with other high-achievers.

It's also the lemming effect... HBS is a rung on the "success" positive reinforcement ladder, like a big funding round. Privileged kids do the "prep school -> ivy -> wall street (quants) / MBB consulting (arts) -> HBS -> PE" track in absence of any other inspiration.


I think this is called failing upwards.

As you allude, an outsider to this cohort crashing into the program should keep expectations in check. Barry Lyndon comes to mind.

I wish it was more socially acceptable to have two year gaps in ones resume and sail around the world or something.


What is the difference between "passport" and "ticket" here?


A ticket allows admission while a passport certifies identity / credentials.

At least in so far as what the words signify, tickets are bought but passports are earned. I think it's a pretty good distinction in this context.


I have trouble believing that a Harvard MBA has had no effect on opportunities or employability, unless you got the one from the continuing education school.


I think OP succumbed to the problem of shifting baselines. The brain adapts and sets the base to whatever your situation is, eventually. You live in a huge house? After a few months you stop noticing it and instead start seeing the small little issues. The rich guy and the poor guy are going to complain equally. Satisfaction and contentment go back to their equilibrium value. Having the Harvard MBA becomes the new normal, life seems equally hard, because you compare with your own new baseline, not with that of other people. Doesn't mean OP can't objectively be correct in his assessment that this MBA did nothing for his employability, because that depends on what he wants to do. But did his goals shift (upwards) too, setting a higher baseline? Seems very possible.


"Salary" and "organizational breadth" are pretty objectively measurable things, and are what I'm using as a measuring stick.

Also: "her"


> "Salary" and "organizational breadth" are pretty objectively measurable things, and are what I'm using as a measuring stick.

That is such a narrow world view. Nor does it change anything I wrote. Nor does you reply make sense, you wrote

> My MBA has had virtually no effect on my opportunities or employability.

That is inconsistent with what you just wrote, something completely different. I'm sure you are aware that humans come up with explanations after decisions? Shown in fMRI scan experiments. I'm not saying this explains what you just did, but it does seem to me that your brain came up with this "explanation" only after discovering my post.


Yes, a lot of people don't understand that MBAs do not carry much weight anymore (if they ever did). I have family members who aggressively pursued such degrees under the assumption that they'd practically guarantee an executive position. I've also met many who've obtained MBAs in an attempt to gain some career advancement, usually to a higher managerial position. I only know of one such person for whom this ploy was successful.

I have to think that the "Harvard" name carries a lot of weight, but MBAs as a class of degrees have, at least over the last ~10 years, been accompanied with a little bit of a snicker, in my admittedly relatively shallow experience with MBA holders. At one point, I did hire a developer with an MBA, but his MBA was a point of concern rather than merit.

>What HBS does well, above all else, is act as a filter. They really do a remarkable job at admissions, putting together a group of folks who are likely (by nature or nurture) to go on to have an outsized impact on the world.

I have to ask. In your opinion, is access to this network really worth the cost of going through the program?


> I have to ask. In your opinion, is access to this network really worth the cost of going through the program?

Not even close to worth it. Networking is pretty easy and the vaunted HBS network is very bloated.

I'd still do it again, but I was coming off an exit and mostly needed a way to keep busy for two years in Boston while my wife did her grad work.


WHy was the MBA a point of concern?


Mainly these two reasons.

1. It signals that one may not be interested in a development career long-term. This is probably the closest you get to the "overqualified" dismissal excuse more commonly given in other fields. The potential impact on the team could be either small or large, depending on the candidate's plan to move himself to a different position.

2. To me, it demonstrated a dependence on formalized schooling processes that may signal an inflexibility of mind that makes it difficult to adapt to the fast-moving world of software. He chose to go through this program not because he was programmed to believe it was the only way he could support himself, which is why most people pursue "higher education", but because he wanted to learn about the business perspective. A strategy that involves getting an MBA is highly inefficient, as discussed elsewhere in this thread.


>remarkable job at admissions, putting together a group of folks who are likely (by nature or nurture) to go on to have an outsized impact on the world.

Outsized impact has very little or nothing to do with personal abilities.

HBS is the ultimate networking opportunity. HBS is way for rich kids to network. On the side it provides networking opportunities for for others who get in. New blood so to speak.


You mention the author confusing correlation with causation and then mention your experience. One data point. Perhaps you are engaged in fallacious reasoning.


My data point is the observation of three years' grads, n > 3000.

Or hell, just the ones I knew well in my year, n=200 or so.


You know for all three years worth of people that the experience wasn't transformative? You know it didn't have an effect on opportunities or employability for these 3000 or so people? I doubt you know all these people that well.

But even if you do know all these people that well and are right about them your conclusion is still not supported. You criticize the author by assuming they are confusing correlation with causation. I suggest you analyze your own reasoning. You are engaged in flawed statistical reasoning and are committing a famous fallacy.

Even if you are absolutely correct about your own experience and the 30000 or so graduates that you know your conclusion could still be wrong. You need much better statistics, or provide evidence that your experience has always been the case for HBS.


I suspect that for a population of 80k with a relatively low variability, n=300 is more than enough.

One of us needs to brush up on their stats...


Indeed one does. One also needs to brush up on argumentative fallacies. It's quite bad to criticize the author's reasoning whilst yourself engaged in bad reasoning.

To be clear, your original post was mostly reasoning along the lines, "My experience was different therefore the author is wrong." And your second post didn't specify why you think your experience was normative. Merely that you think in a band of a few years you know how others benefitted or didn't from HBS. This is especially bad reasoning as the article talked about HBS influence over a much broader range of years.


Prof. Michael Jensen's higly-cited 1976 paper [0] identified the principal-agent problem [1] then argued that a company's sole purpose is to maximize shareholder value. This thinking is what gradually led to the now widespread obsession with stock prices and short-term results.

Short-termism is harmful.

Jack Welch said it best in a Financial Times [1] interview:

“On the face of it, shareholder value is the dumbest idea in the world. Shareholder value is a result, not a strategy… your main constituencies are your employees, your customers and your products. Managers and investors should not set share price increases as their overarching goal. … Short-term profits should be allied with an increase in the long-term value of a company.”

[0] Theory of the firm: Managerial behavior, agency costs and ownership structure, http://www.sciencedirect.com/science/article/pii/0304405X769...

[1] https://en.wikipedia.org/wiki/Principal%E2%80%93agent_proble...

[2] http://www.ft.com/intl/cms/s/0/294ff1f2-0f27-11de-ba10-00007...


> [Professor Michael Jensen] advocated an "agency" theory of management in which management's sole duty was to maximize shareholder value. This upended the long-held "stakeholder" model, in which management was seen as having broader obligations to a corporation’s workers, customers and communities.

It's interesting how that theory has filtered into public consciousness. My sense is that the public doesn't grasp it as one theory among many possibilities, but as a law of nature (a pattern I see for many ideas - people lack perspective that the theory is one of many possibilities, and different concepts were popular before and others will be popular later). You can read people using agency theory, thought not by that name, on HN regularly.

Consider how amazing that is: An academic comes up with a theory, and eventually tens of millions (or more) are repeating it. Is it marketed? Even if it is, many or most of the people repeating it probably have no idea of the marketing. Is it part of an intentional political propaganda campaign? I know other popular ideas successfully pushed on the public this way (one I know of in particular is the "death tax", and I've observed others where the talking points frame and effectively end the discussion).

What incredible power. Is it random? Can it be directed and controlled? Who knows how and who is doing it? I know political groups do it to an extent, as I said, but I don't know their range or accuracy, so to speak.


Nautilus had a great article a while back on how things get popular[1]: it's more a function of social networks than of the things themselves. The research detailed in the article looked at music specifically, but I think the idea generalizes to most anything including hard-to-falsify academic theories.

http://nautil.us/issue/5/fame/homo-narrativus-and-the-troubl...


That last paragraph is a dangerous rabbit hole to go down


Clayton Christensen, Harvard Business School prof, has promoted a tax structure that has much lower taxes on investments that last multiple years (say 5), and higher taxes on short-term investment (less than a year) as a way to incentivize long-term R&D. I like this as a mechanism to change the way MBAs think about profit and investment. I.e., as something that I'd love to see more at Harvard get behind.


I'm not sure that would help much. I think the risks and uncertainties of long-term R&D increase much faster than any tax incentive can compensate for. My favorite story is the (real) history of Silicon Valley. I once thought that was a proof for what private enterprise can accomplish. Turns out that SV is actually based on the unrestrained spending on R&D (of borrowed money, of course) by government during WWII. [0]

Private enterprise wants a level of certainty, long-term or not. Tax incentives don't reduce the uncertainty. I doubt that money is the major constraint. Look at some major corporations basically "swimming" in money, and they also should be able to borrow easily (and cheaply) at this time.

[0] https://www.youtube.com/watch?v=ZTC_RxWN_xo


I am admittedly out of my depth here, but I want to make a quick comment about something I've noticed from HBR publications (which claim to be primarily from HBS professors).

Typically, they strike me as the most stereotypically "academic" writings that I regularly run across. Meaning they are working hard to maintain an appearance of rigor and neutrality, but spend a great deal of time and effort squinting and straining to produce an outcome that aligns with their intended philosophy and desired conclusions. There is just so much output that is naive and tone-deaf to what's experienced on the ground in the white-collar world. While you can arrange some battery of numbers to support almost any case ("there are lies, damned lies, and statistics"), it doesn't change the efficacy or practicality of some of the things they end up endorsing/suggesting.

This is not a universal or blanket condemnation, as I do follow HBR precisely because sometimes their output is interesting or valuable. But many times it's clearly the result of a bubble.

One of the most offensive habits of academia is to elevate people to professorships almost straight out of school, despite these people having virtually zero real experience practicing the art they claim to know how to teach. This is particularly offensive when it comes to fields where the art is a crucial part of the application, like business and law -- which, afaik, are two fields where these farcical elevations happen routinely.


What do people really think about business school degrees? To me there seems to be a large disconnect between what we think of the person who takes one (ambitious, hard-working) and what he is taught there (basically nothing).

I went to business school as part of my degree at a top UK institution. They have 1 year MBAs so the joint school people got a fair taste of what the MBAs did.

And let me tell you, as half engineers nobody was impressed. Is this really what they do? Read a bunch of stories about Betamax or various Japanese firms, and extract lessons from that? That's not even a rigorous history course.

As to what the overall effect is, it just appears there is a class of firms who are fearful of hiring the wrong people, so they hire a brand name. These people are more likely to get into such firms, but it's not clear at all anything from the course is used there in a way that is more than superficial.

Have they damaged society by being too focused on short term gains? Well, it's not just Harvard and MBAs that went that way in the 1980s. Society as a whole did, too.


Several friends who have done MBAs have all told me the same thing: When you do an MBA, you pay for two things. First, for the network of alumni. Second, for the badge in your resume. In both cases, obviously the more prestigious the school, the better (and more expensive). Since that's what would carry more weight as a brand name (i.e. Harvard) and also richer / more influential people will attend as alumni, thus those personal connections being more valuable. Recently I heard about some MBAs called "Executive MBA" or something like that, which were basically shorter (1 year) and more expensive. They justified this by saying they were targeted at people already on the workforce, usually in management, who wanted to move up, and would really appreciate 1 year titles instead of 2 years. But again, it's for the same reason. More expensive title => "Wealthier" alumni => Looks better on your resume. And to top it off, you just have to spend a year "studying".


This almost sounds like a finishing school for corporate courtiers.


I always got the impression that MBAs were about signalling that someone was serious about wanting to be be in the "management" stream (or if I'm feeling cynical "class") rather than a mere practitioner in a particular area.

Edit: Nobody that I know who did an MBA seemed that interested in the area they worked in. I'd rather do a PhD in History or Archaeology than an MBA - perhaps my own desire to signal that I have no interest in "management" for its own sake.


> I'd rather do a PhD in History or Archaeology than an MBA

To be fair, business schools offer PhDs which are different beasts I suspect.


What underlines this, I think at least, is the fundamental failure of representative democracy: if the representatives can gain larger personal wealth especially on the long run by electing rules that concentrate wealth they will do so. This will work, alas, even without any corruption -- when they leave their post they will be able to get compensated for participating in the "system" and they know this. Because a Business School can only corrupt society if regulation lets that happen.

Breaking this up is not easy. I believe liquid (or delegative) democracy is the road forward. It is very interesting that the early Soviets before the Bolsheviks took over were pretty much this. http://www.ditext.com/voline/89.html But the time was not right for that -- now everyone has the means to communicate in their pocket so perhaps it's time now.


It's so easy to criticize Harvard Business School (HBS). Outsiders -- and insiders -- have been doing it for decades now, this book being the latest example. The Web is littered with accounts of spoiled Harvard MBAs full of hot air who have done stupid things.

Yet the school somehow continues to churn out an impressive number of people who rise to the top in extremely competitive fields, from Michael Bloomberg to Ray Dalio to Mitt Romney to Sheryl Sandberg, to name just a few.[1]

How does THAT happen?

My take is that over time HBS has evolved a process for identifying and attracting a disproportionate number of people with that kind of potential. Speaking from personal experience, I know and/or interact regularly with a good number of HBS grads. While some of them are indeed full of hot air, a surprising number are truly impressive individuals who are not only thoughtful and likable, but also capable of "running through walls" to achieve their goals.

[1] https://en.wikipedia.org/wiki/Harvard_Business_School#Notabl...


Or it allows you to hobnob with extremely wealthy people and make connections with them.


What does a university do with a $32.7 billion endowment?


Invests it, and runs the university on the returns. Assuming a 4% yearly rate of return, then that's around $1.5B a year in money available for operating expenses. As a point of comparison, Facebook spends ~10x that (~$15B), Uber spends ~2x that ($3B), etc. Harvard obviously has additional revenue streams, but there you go.


Comparing tech firm operational costs to universities is a bit ridiculous -- Facebook runs multiple private datacenters globally. Most universities get by with a dozen clustered on campus serving administrative, educational and research objectives. It also serves more or less the planet, whereas the Harvards and Berkeleys of the world exist to serve a few dozen thousand at at time.

$1.5B pays for quite a bit of research. Or, more accurately, a lot of real estate speculation for the board of trustees to play.


Isn't most research funded out of external industry and government grants that also pay into the university (ie grant overhead)?

Given that even most top research universities get by on a fraction of Harvard's endowment—while running largely comparable research programs—it's definitely possible to do without. Harvard obviously does strong research, but not that much stronger than the rest of the top 100 (top 200? 500?) universities. Certainly not 10–50× stronger!

My impression is that what really makes Harvard and other highly endowed universities stand out is their really aggressive financial aid. That's probably one of the things that's almost entirely funded out of the endowment as opposed to external sources or tuition and fees.


Of course! The point was to put the number next to a number known by the typical HN reader, not to imply the businesses were at all similar.

> Most universities get by with a dozen clustered on campus serving administrative, educational and research objectives

... A dozen what?


> > Most universities get by with a dozen clustered on campus serving administrative, educational and research objectives > > ... A dozen what?

Looks like GP meant "a dozen data-centers", to compare the costs.


I figured that, but I wanted to give the chance at clarification, because we were mainly discussing operational expenses, and there's no way data centers even clears 1% of the cost of running Harvard


It's mostly tied up in investments.



On the same note, here's some research on the impact of Harvard MBAs as CEOs: http://www.mintzberg.org/blog/mbas-as-ceos

> "A decade after its publication in 1990, I looked at a book called Inside the Harvard Business School, by David Ewing, long an insider. (The first line was “The Harvard Business School is probably the most powerful private institution in the world.”) The book listed 19 Harvard alumni who “had made it to the top”—the school’s superstars as of 1990. My attention was drawn to a few of them who would not have been on that list after 1990 [...] So Joseph Lampel and I studied the post-1990 records of all 19. How did they do? In a word, badly."


It was explained to me once that it wasn't Harvard Business School that was responsible for the decline of capitalism, but Wharton. That the two represented two very different approaches to people. Harvard supposedly taught to value employees and customers as people, and to strive for social good as a primary goal. Wharton supposedly taught to treat employees and customers as renewable resources, dehumanizing them, and to strive for profits and shareholder value as the only goals. The source was biased to old-school ivy league, but someone I trust. I've never attended either school.


Wharton is old-school Ivy League. You probably know this, but just from the way you worded it, wasn't clear.


I'm sympathetic to arguments that MBAs, particularly at elite institutions, mostly benefit the graduate and not the company that employs him. I read an interesting study recently that made this claim by comparing the stock-returns of companies led by Harvard MBAs vs the market as a whole. Sure, the study is no smoking gun, but it's on the right track.

In contrast, the entire book reads like a series of anecdotes. And given that Harvard-MBA recruits the best and most ambitious students who go on to occupy numerous positions of power, you can always find a harvard MBA scapegoat behind every business failure that ever occurs. I don't see how this proves anything about the quality of Harvard-MBAs, other than the fact that they aren't perfect supermen. One could use this exact same line of reasoning and "evidence" in order to claim that white men are the cause of all problems in America.


HBS now has courses like Building Sustainable Cities and Reimagining Capitalism, both courses created in response to what I presume is student interest in subjects beyond the profit maximization motive.

Outside of a few courses and a few class discussions within each course that are centered in exploring social value, the overwhelming majority of the experience requires students to check their social mindsets at the door and think like stewards of shareholder value.

One of the saddest parts about the HBS classroom is its reliance on student participation and student-led learning. While this is an effective way of training students to assert their opinions, I have seen several circumstances in which groupthink and conformist pressure keeps the "social" viewpoint at the margins of the in-class discussion. HBS just isn't the place for critically thinking about anything other than profit and loss.

Edit: grammar


These kinds of narratives are so appealing, but they are so very difficult to justify.

Harvard has monumental cachet, and 2008 is still an interesting topic. So it makes for a super interesting argument to suggest that Harvard is somehow responsible for 2008. But come on. This is simplistic, to say the very least.

This kind of narrative reminds me of when people "explain" why Apple under Steve Jobs was so successful. It's safe to theorize in that case because we can't rerun history and do experiments. 2008 is the same way.


It does seem like Jensen was key in evangelizing ideas that tied executive compensation to stock performance. This even led to a 1993 IRS rule [1]. Here "value" was clearly shareholder value; the dollar value of stock.

Then, in 2000, on the verge of the Enron collapse, Jensen offered an updated theory - "Value Maximization and the Corporate Objective Function" [2]. Here, he suggested that the definition of "value" perhaps needed to evolve, and consider stakeholders. However, I'm not sure this later work had the impact of changing executive compensation, or corporate decision-making, probably just created debate.

Probably feels something like not being able to fix a bug in production..

"Undeniably one of the most influential business theorists of modern times, he (Michael Jensen) advocated an “agency” theory of management in which management’s sole duty was to maximize shareholder value. This upended the long-held “stakeholder” model, in which management was seen as having broader obligations to a corporation’s workers, customers and communities."

Wikipedia [1]: "After Jensen and Murphy (1990), Congress passed a law, making it cost effective to pay executives in equity. As a result, executives had a financial incentive to focus their efforts on increasing stock price. In the short run, some executives even manipulated accounting numbers (e.g., Enron, Global Crossing) to achieve this goal"

HBS '00 [3]: Jensen, new theory of Enlightened Value Maximization: "We must give people enough structure to understand what maximizing value means so that they can be guided by it and therefore have a chance to actually achieve it. They must be turned on by the vision or the strategy in the sense that it taps into some human desire or passion of their own—for example, a desire to build the world’s best automobile or to create a film or play that will move people for centuries. All this can be not only consistent with value seeking, but a major contributor to it.

And this brings us up against the limits of value maximization per se. Value seeking tells an organization and its participants how their success in achieving a vision or in implementing a strategy will be assessed. But value maximizing or value seeking says nothing about how to create a superior vision or strategy. Nor does it tell employees or managers how to find or establish initiatives or ventures that create value. It only tells them how we will measure success in their activity."

[1] https://en.wikipedia.org/wiki/Michael_C._Jensen [2] http://www.hbs.edu/faculty/Publication%20Files/00-058_f2896b... [3] http://hbswk.hbs.edu/item/value-maximization-and-stakeholder...


"France’s prestigious “Grandes Écoles” — not H.B.S. — appear to have generated a disproportionate number of the financial engineers who unwittingly helped cause the crisis.)". Wow... just wow.


Read Chomsky and weep.


I'll read Chomsky's work as a linguist. As for his economic and political writings, I'm going to pretend that they don't exist until he pulls his head out of his ass and stops excusing genocide.


I thoroughly enjoy John Ralston Saul -- he popped up on HN a while back[1]. A Canadian philospher with a GSOH - what's not to love?

From his Doubter's Companion (highly recommended) - an occasionally acerbic description of many things - comes this definitive entry under Chicago / Harvard School of Economics:

> A great centre of contemporary SCHOLASTICISM. The economists working there and produced by it are as important to the stagnation of useful thought as the Schoolmen of the University of Paris were at the height of the Middle Ages.

> Like that of the Paris scholastics, their mastery of highly complex rhetorical details obscures a great void at the centre of their argument. They also share a tactical genius for exporting their conceptual definitions to less important centres around the world. The result is a pleasing symphony of international echoes imitating their calculations and cadences and so confirming their correctness, even when their policies bring economic disaster. The percussion section of Chicago’s orchestra is the Nobel committee for economics. Each golden medal is like another congratulatory parchment presented at the end of an elaborate theological debate.

> But what of content? There isn’t much. What of Friedrich Hayek and Milton Friedman? These minor Thomists preach little more than inevitability and so counsel passivity.

> What they call libertarian economics is a remarkable revenge of the scholastics on the men of the Enlightenment, who had theoretically destroyed them. Peel away the tangle of intellectual leaves from the Chicago School and what remains is a great clockmaker god who has set the world ticking. But the conclusion of the Enlightenment was that god’s indifference left humans free to organize the world as they wished. Chicago has so deformed this idea as to invert it. The great clock has been turned into an absolute, all-encompassing system. Better than an ideology, the world is its own absolute economic truth. We must remain passive before its majesty.

> This is a denial of Western experience. It is nonsense which simply comforts the power slipping increasingly into the corporatist structures.

> Strategic thinking can save a great deal of time wasted over tactics. A large number of America’s economic problems, and those of the West, could be solved by shutting down the Chicago School of Economics.

> This would not prevent the academics employed there from preaching their essentially anti-social and amoral doctrines. They would be gathered up with delight by the hundreds of imitation Chicago Schools. The purpose of closure would be simply to disentangle a tendentious ideology from its unassailable position within contemporary power structures. The same sort of liberating shock treatment was applied to European civilization in 1723 when the Society of Jesus (Jesuits) was disbanded. The effect was to set free the ideas of the Enlightenment.

[1] https://news.ycombinator.com/item?id=10139144


one solution for the "...the Moral Failure of the MBA Elite " - part:

https://en.wikipedia.org/wiki/MBA_Oath

"MBA Oath (short version)"

"As a manager, my purpose is to serve the greater good by bringing people and resources together to create value that no single individual can create alone. Therefore I will seek a course that enhances the value my enterprise can create for society over the long term. I recognize my decisions can have far-reaching consequences that affect the well-being of individuals inside and outside my enterprise, today and in the future. As I reconcile the interests of different constituencies, I will face choices that are not easy for me and others.

Therefore I promise:

-I will act with utmost integrity and pursue my work in an ethical manner.

-I will safeguard the interests of my shareholders, co-workers, customers and the society in which we operate.

-I will manage my enterprise in good faith, guarding against decisions and behavior that advance my own narrow ambitions but harm the enterprise and the societies it serves.

-I will understand and uphold, both in letter and in spirit, the laws and contracts governing my own conduct and that of my enterprise.

-I will take responsibility for my actions, and I will represent the performance and risks of my enterprise accurately and honestly.

-I will develop both myself and other managers under my supervision so that the profession continues to grow and contribute to the well-being of society.

-I will strive to create sustainable economic, social, and environmental prosperity worldwide.

-I will be accountable to my peers and they will be accountable to me for living by this oath.

-This oath I make freely, and upon my honor."

more:

- http://mbaoath.org/

- https://hbr.org/2009/06/why-we-created-the-mba-oath

- FT:"Is the MBA oath still relevant?" 2015 https://www.ft.com/content/e853f5a2-fe41-11e4-8efb-00144feab...

- debate: "Why MBAs should not sign the Harvard Business School oath"? http://knowledge.insead.edu/leadership-management/why-mbas-s...

or alterative suggestions from the "debate":

"I pledge to maximise the wealth of the people who pay my salary, i.e. the shareholders, unless the shareholders tell me in advance that they want me to do something else. I will do my best to learn how to do this by taking the relevant courses"


Oaths don't really mean much if there's no recourse to not following them


The failing New York Times failed to point out the author of the critique, James Stewart, is a Harvard grad himself. Hardly an unbiased article. No conflict-of-interest disclosure. Many half-truths and disingenuous statements in defense of Harvard.

Sad.


> junk-bond-induced takeover mania and resulting scandals of the 1980s; the corporate scandals of the 2000s; the egregious increase in the pay gap between chief executives and ordinary employees; the real estate mortgage bubble and ensuing financial crisis; even the election of Donald Trump

Just leave out the part about President Trump. Plenty of people like him.

> Harvard has obsessively pursued money, sending a disproportionate number of its graduates to consulting firms

I assume the students are the ones pursuing money

> An obsession with stock prices and short-term results was the motive for many of the accounting scandals of the 2000s

Was it ? I thought it was unscrupulous people doing accounting fraud and information asymmetry between shareholders and executives.

> McDonald insists he isn’t “anti-business-school” (he attended Wharton), nor is he “anti-wealth.

But just a few sentences above :

> McDonald notes that in 1992, the C.E.O.s of Fortune 500 firms made an average of $2.7 million a year. By 2000, the average was $14 million.

> It’s a shame that some of them didn’t see the looming disaster and sound an alarm

Some did and made a lot of money from shorting these bonds. Only when it became possible to short these bonds that the bubble finally collapsed and the money went to fund more useful ventures.

> lay a part in helping more people who think about business rediscover a purpose other than profit

A better solution would be to ensure more information symmetry ( even through government action ) and having more ambitious people in regulatory agencies. 'Pitting ambition against ambition' ( to use a Madison quote ) is a system that has historically worked very well to curb these types of predatory behavior




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: