Monday, June 5, 2023

Stephens at Chicago; effective organizations

Brett Stephens gave a great commencement speech (NYT link, HT Luis Garicano) at the University of Chicago. One part stood out to me, and worthy of comment. Bret starts with the problem of Groupthink:

Why did nobody at Facebook — sorry, Meta — stop Mark Zuckerberg from going all in on the Metaverse, possibly the worst business idea since New Coke? Why were the economists and governors at the Federal Reserve so confident that interest rates could remain at rock bottom for years without running a serious risk of inflation? Why did the C.I.A. believe that the government of Afghanistan could hold out against the Taliban for months but that the government of Ukraine would fold to the Russian Army in days? Why were so few people on Wall Street betting against the housing market in 2007? Why were so many officials and highly qualified analysts so adamant that Saddam Hussein had weapons of mass destruction? Why were so many people convinced that overpopulation was going to lead to catastrophic food shortages, and that the only sensible answers were a one-child policy and forced sterilizations?

Oh, and why did so many major polling firms fail to predict Donald Trump’s victory in 2016?

Conspicuous institutional failures are the question of our age We could add the SBV regulatory fiasco, the 2007 financial regulatory failure, the CDC FDA and numerous governments under Covid,  and many more.  Systemic incompetence doesn't just include disasters, but ongoing wounds from the Jones act to California's billions wasted on obviously ineffective homeless spending. 

The list is a bit unfair, of course. Selection bias: These are the grand failures, but large organizations occasionally produce some successes. For every Metaverse there is an iPhone, which I certainly thought a dumb idea at the time. And it's always easy to see idiocy with hindsight, but it's a lot harder in real time. De-growthing our economies and spending trillions in the name of carbon reduction will be seen, 20 years from now, either as a farsighted visionary move that saved civilization, or a grand collective delusion. Which is it? Who is the naked emperor and who is the little girl on the sidelines of the parade? Remember too that the gadflies are usually wrong. 

But the question on my mind is this: How do you structure large organizations to avoid such catastrophic mistakes? As an economist, and a macroeconomist at that, it's something I don't know anywhere near enough about. 


... Why is it that, when you bring together a lot of smart people in a room, their collective intelligence tends to go down, not up? Why do they always seem to press the mute button on their critical faculties when confronted with propositions that, as an old colleague of mine liked to say, ought to vanish in the presence of thought? 

It's not obvious people's critical faculties are impaired, but their incentives to speak out about them are. 

First, the problem isn’t that people aren’t smart. It’s that they are scared.

To yell stop when everyone else says go — or go when everyone else says stop — takes guts, and guts aren’t part of any kind of normal college curriculum. In my generation, the hardest people to say “no” to were the people who had professional power over us. In your generation, I think, it’s the people who are in your own ideological tribe. Whatever it is, how many of us, if we’re honest with ourselves, really have that kind of courage?

Second, there is the problem of rationalization — of smart people convincing themselves, and others, of some truly dumb things.

Robert McNamara, one of the original “Whiz Kids” and probably one of the brighter bulbs in 20th-century American public life, was one of the fathers of the Vietnam War when he was at the Pentagon, and of the Third World debt crisis when he was at the World Bank. Somehow, he always managed to convince the other smart people in the room that he was right. Will you be able to notice the underlying flaw in an idea when the arguments for it sound so persuasive?

 Or, he convinced them to silence their doubts and go along. 

Third, there is the psychological dimension.

Some people are inveterate truth seekers. They are almost congenitally willing to risk rejection, ostracism, even hatred for the sake of being right. But most people just want to belong, and the most essential elements of belonging are agreeing and conforming. ...the usual emotional companion to intellectual independence isn’t pride or self-confidence. It’s loneliness and sometimes crippling self-doubt.

This is insightful, but it's not getting us to the question on my mind: Why do some institutions seem more prone to groupthink disasters than others? Bret's final insight gets to that: 

here’s a fourth factor, maybe the most crucial. It’s culture. Does the culture of a society, or of an institution, encourage us to stand out or to fit in; to speak up or to bury our doubts? Does it serve as a conduit to groupthink, or as an obstacle to it?

I mentioned a moment ago that all of us like to think of ourselves as independent thinkers, even if comparatively few of us really are. There’s an institutional corollary. Nearly every American institution outside of certain religious orders claims to encourage open debate and — that awful cliché — thinking “outside the box.” Apple’s famous slogan, “Think Different,” was one of the most successful ad campaigns of my lifetime...

But, at least in my experience, very few institutions truly welcome it, at least when it exposes them to any sort of pressure or criticism, much less loss of social capital or potential revenue...

But this doesn’t always have to be the case. Institutions can, in fact, practice what they preach. They can declare principles, set a tone, announce norms and expectations — and then live up to their principles through regular practice. They can explain to every incoming class of students or new employees that they champion independent thinking and free expression in both word and deed. They can prove that they won’t cave to outrage mobs and other forms of public pressure, either by canceling invited speakers or by never inviting controversial speakers in the first place.

There’s a way this is done. It’s called leadership. You have one magnificent example of it right on this stage, in the person of John Boyer. And you have had a historic example of it in the person of Bob Zimmer. I want to say a few words about him.

That's as far as Bret goes, appropriately for a graduation speech at Chicago. So we have one answer to my question: Some institutions have cultures that welcome emperor is naked commentary, and most do not. Leaders can set cultures. 

I think this just scratches the surface. A college's free speech culture is nowhere near as consequential as a government making a decision to go to war, or any of Bret's other examples. Institutions eventually have to have mechanisms for coming to a decision, closing ranks and pursuing it. If you're going to go to war or not, you have to make a decision and not keep arguing about it forever. If you've ever participated in any group decision you know there are gadflies bringing up stupid points over and over, and if you have too much discussion you're never going to get anywhere. I think institutions in today's government are in CYA mode for good political reasons. The Fed doesn't have a groupthink culture because it wants to, but because in today's Washington admitting mistakes would lead to a completely ineffective institution under constant attack. Again, the gadflys are also mostly wrong too! 

I do think there are additional institutional structures that could help to promote good decision making. An official devil's advocate to big decisions, and making sure that isn't a career dead end is one useful concept I've heard of. But the larger question of just what those are remains something I'd like to know more about. 


  1. Two points: (1) we are critiquing a realized path.The outcome is not based on one decision but series of decisions. These decisions are taken by multiple stakeholders. Even if everyone's incentives are aligned with the vision and mission of the Institution, these decisions are somewhat relatively unique to the decision maker and the time when they are taken.
    (2) I suppose academia allows for dissent and that is why monetary theory and fiscal theory of price level co-exist ? Could that be the reason, why we have segregated industry and academic world?:

  2. Thanks John this sounded like a thought-provoking talk: I wish more commencement speeches were like this. This topic of what makes a wise organization connects to what makes a crowd wise. James Surowiecki's The Wisdom of Crowds argued for four conditions: (i) diversity; (ii) independent decisions; (iii) the problem involves decentralized knowledge; (iv) there is an aggregation mechanism. He claims, largely through anecdotes that (i) and (ii) are the best remedies to the groupthink you highlight. So perhaps one goal of the organization is to shape who is in the room making decisions as well avoiding too much interactions. It would make for an interesting lab or field experiment to see of these conditions are necessary and sufficient.

  3. Just ruminating, but isn't this part of what Ronald Coase asked himself in his Theory of the Firm paper (written as an undergrad, BTW), and what Oliver Williamson dedicated his academic life to studying?

  4. The basic reality is that every distribution has two tails. You can't endlessly bemoan the grand failures of society or large corporations unless you are willing to give up let's see here - the telephone, the automobile, the PC and on the institutional side - Democracy. All of these inventions were dismissed as crazy ideas too!

    There is something to ensuring that organizations have effective channels for challenge and debate but that should not be confused with "stopping all bad ideas" because the bad ideas are simply the cost of the really great ones. If you truly would like no more grand failures then go sit in a cave - the process of growth brings failure as a necessary byproduct.

  5. Re effective organizations. Some favorite aphorisms come to mind:
    • I am reminded of a line spoken by Supt. Reginald Bright in a scene in the Masterpiece Theatre series “Endeavour,” where, just before leading his police officers into a deadly confrontation with criminals that include a corrupt colleague, he says, “A moment of courage or a lifetime of regret.”
    • Without doubt, the qualities of guardedness, heedfulness and wariness are necessary in the conduct of battle, but it is a postulate that should be accompanied by the succinct response of Cardinal Mercier, Primate of Belgium, who became a hero to the world for his defense of Belgium during its sufferings after the German invasion in World War II: “When prudence is everywhere, courage is nowhere.”
    • “Outs” in institutions prize independence, even as the “Ins” manifest fear of being in the losing coalition.
    • "Easy to enlist a thousand soldiers, but, ah, one general." —Chinese proverb
    • If you lose your wealth, you've lost nothing; if you lose your health, you've lost something; if you lose your character, you've lost all. —Billy Graham

  6. It is totally about leadership. I’m not sure how to build structures, beyond leaders visibly supporting risk taking and speaking truth.

    I spent about 25 years in very large scale computer sales. About 5 at IBM first and about 13 at Sun Microsystems with a few years at other places. IBM had great training, especially for beginners and salesman right out of college, but the culture was very paternalistic and top down and very into managing appearances. A great company in a lot of ways, but clearly the glory days were in the 50’s and 60’s, and by the 80’s when I was there starting to fade.

    Anyway, when I joined Sun in 1995, I was very struck by my New Field Employee training. By definition new people who represent the company to customers. The very first item on the week’s agenda was “Chairman’s Message”. Based on my years at IBM I was expecting to sleep through a video, but at 8:30 in the morning, Scott McNealy the CEO, Chairman and Founder comes walking down the aisle to spend an hour speaking with about 20 of us. Most striking about his remarks was “I’ve always believed that to ask permission is to seek denial. You should feel that way too.” An absolutely inconceivable remark by anyone in IBM leadership. I clearly remember it nearly 30 years later.

    It was also quite memorable that Scott also said “We are a Product company.” The last item on the agenda that day was a talk with Ed Zander the President of Sun. (Incidentally, that talk included beers which would also have been inconceivable at an IBM meeting). Anyway, Ed said “Sun is a services company”. Feeling bullet proof (and not having a name badge on) i raised my hand and said “this morning Scott stood right there and said We are a products company to which Ed said “Scott said that again?!” And we moved on. Could not have been more clear that it was ok to disagree and speak out and make your own decisions. Sun in the 90’s was a great company and very much the best place I ever worked.

    In my view, what really killed Sun was the .com bust - Sun was riding high and doing incredibly well, very open place, when the whole market fell apart. That put so much stress on the organization and instilled so much fear that things were never the same again. Scott remained in charge and many things stayed the same, but there was just a lot of fear and concern and that colored everything. Not punishment or action by leadership, just generalized fear had a huge impact on the culture.

  7. Re predicting Trump’s 2016 victory: how would one ”predict” an event that election modelling (based on running thousands of simulations) assessed had a probability in the order of 0.25? To my recollection, there were quite a few people and organisations (incl. Nate Silver’s) who emphasized that a Trump victory can by no means be ruled out. I do not think that there was anyone that could have ”predicted” Trump winning in the sense that they would have had rational grounds to conclude that a Trump victory was inevitable.

    1. Re. the problems with such models, see:

  8. "I do think there are additional institutional structures that could help to promote good decision making. An official devil's advocate to big decisions, and making sure that isn't a career dead end is one useful concept I've heard of. But the larger question of just what those are remains something I'd like to know more about." --JHC.

    Ans.: The "ask" is too large. Break it done into its constituent parts. Which is to say, "Rome wasn't built in a day."

    For example, consistent with the focus of this blog site over the past year and a half, consider the U.S. government's projected deficits and the knock-on effects that those will have going forward into the 2030s. For "institutional structures" that fulfill the "devil's advocate" function, look no further than the Congressional Budget Office and its estimates of the negative effects of deficit-spending on the share of the federal government budget expected to go to debt servicing in near future years. The impact is not difficult to imagine, but the consequences are uncertain. -- Does the U.S. dollar remain the world's reserve currency; can the U.S. meet its obligations on a ballooning debt and its promises to its retirees, and still remain competitive as an economy? Or, does it follow Argentina's example?

    Consider what the U.S. security system (defense) capability will be when the majority of spending is directed to making sure that the debt is properly serviced. Under what scenarios would it be appropriate to abandon Social Security in order to ensure that the U.S. maintains its capability to defend its interests at home and abroad?

    By isolating individual components, in order to address each adequately, there is created a need to provide for integration of the components to achieve some near optimal condition for the whole (is it greater than the sum of the parts?) And, decisions as to priority of effort also require attention. The first step is to establish a framework, a division of labor, so to speak. Then tackle each element individually, and then as a component of the whole; followed by integration to sum the parts, etc. Then iterate until the marginal benefit of further effort falls below the marginal cost of undertaking that further effort.

    A caution -- before commencing, consider whether the outcome is achievable with the social structure currently in place. Who will co-operate, and who will sabotage the undertaking. The first question to address, then, must be, “Is it worth saving?” The follow-on questions, if the first is answered in the affirmative, are “What are we willing to pay?”, and “What are we willing to accept?”. The answers to those two questions will determine whether there is a will and what the limits of sacrifice are expected to be, and how determined the majority are to right a listing ship of state.

    There is some indication that conditions are developing which would facilitate such an undertaking, but it is at present tentative and uncertain. Best of luck to you.

  9. William WiltschkoJune 6, 2023 at 2:26 PM

    The values of Intel Corporation for its first thirty years were those of Andy Grove. One important value was "constructive confrontation", which Andy practiced, and encouraged. If you had a contrary opinion, you were expected to air it and defend it. The value made everyone uncomfortable, but issues were often thrashed out quickly and thoroughly because we knew "only the paranoid survive", the title of one of his books.
    Today, Intel has lost some of these values, but Pat Gelsinger, the current CEO who grew up under Andy, has stated a desire to return to "Grovian" values. In a business where you bet the company every few years, anything less than constructive confrontation is suicidal. Yes, leadership is important, especially choosing the right, unique leader.

  10. In order to judge whether the FOMC misjudged the rate of inflation relative to target and thereby qualifies under the heading "institutional failure", first determine if this constitutes a failure in execution of policy. To that end, examine the 8/27/20 policy statement. If the FOMC is executing in conformance to policy, no institutional failure occurred. Instead, misapprehension of the FOMC’s policy is likely.

    "2020 Statement on Longer-Run Goals and Monetary Policy Strategy." [Abridged]
    Adopted eff. 1/24/2012; as amended eff. 8/27/2020; reaffirmed eff. 1/31/23.

    Firmly committed to fulfilling its mandate of promoting: • maximum employment, • stable prices, and • moderate long-term interest rates.

    Seeks to explain its monetary policy decisions as clearly as possible. Such clarity facilitates: • well-informed decision making by households and businesses, • reduces economic and financial uncertainty, • increases the effectiveness of monetary policy, and • enhances transparency and accountability, which are essential in a democratic society.

    Employment, inflation, and long-term interest rates fluctuate over time in response to economic and financial disturbances.

    Monetary policy plays an important role in stabilizing the economy in response to these disturbances.

    The primary means of adjusting the stance of monetary policy is through changes in the target range for the federal funds rate.

    The maximum level of employment is: • is not directly measurable and • changes over time owing largely to nonmonetary factors that affect the structure and dynamics of the labor market.

    Consequently, it would not be appropriate: • to specify a fixed goal for employment; • rather, policy decisions must be informed by: assessments of the shortfalls of employment from its maximum level, recognizing that such assessments are: necessarily uncertain, and subject to revision.

    The inflation rate over the longer run is: • primarily determined by monetary policy, and hence the FOMC has the ability to specify a longer-run goal for inflation.

    Reaffirms: • its judgment that inflation at the rate of 2 percent, as measured by the annual change in the price index for personal consumption expenditures, is most consistent over the longer run with the Federal Reserve's statutory mandate.

    Judges that longer-term inflation expectations that are well anchored at 2 percent foster: • price stability and • moderate long-term interest rates and • enhance the ability to promote maximum employment in the face of significant economic disturbances.

    In order to anchor longer-term inflation expectations at this level, the [ ] seeks: • to achieve inflation that averages 2 percent over time, and therefore judges that, • following periods when inflation has been running persistently below 2 percent, appropriate monetary policy will likely aim: to achieve inflation moderately above 2 percent for some time.

    Monetary policy actions tend to influence: • economic activity, • employment, and • prices with a lag.

    In setting monetary policy, the [FOMC] seeks over time to: • mitigate shortfalls of employment from the its assessment of its maximum level and deviations of inflation from its longer-run goal.

    Moreover, sustainably achieving maximum employment and price stability depends on a stable financial system.

    Therefore, policy decisions reflect: • longer-run goals, • its medium-term outlook, and • its assessments of the balance of risks, including risks to the financial system that could impede the attainment of the goals.

    The employment and inflation objectives are generally complementary. However, under circumstances in which it judges that the objectives are not complementary, it takes into account: • the employment shortfalls and • inflation deviations and • the potentially different time horizons over which employment and inflation are projected to return to levels judged consistent with its mandate."

  11. 1) Mainstream Media are composed of people who vote for one political party, get all their news fed to them and there is not one "reporter" left worth a damn.
    2) Censorship by the media/government complex is on an industrial level.
    3) 99% of the bureaucrats in Washington DC are Democrats.
    4) 95% of all college professors are Democrats
    5) 95% of all Public School Parasites are Democrats
    6) Not one CEO of a Fortune 100 Company donated to the Trump campaign.
    7) 100% of the Pentagon General Staff are there because they are 100% Woke.

    and there is Groupthink? Well, I'll be damned.

  12. I am reminded of the work of Jensen and Meckling regarding decision making in organizations. They argue (related to Hayek's work) that the key to good decision making is to unite incentives and knowledge, i.e., having decision makers who are knowledgeable and properly incentivized. Big organizations, or committees in big organizations, seem to fail most often regarding the incentives. Smart people are all around, but put them in a big group with "shared governance" and the incentives often go by the wayside. No one is "responsible" so you get all a lot of what Stephens describes; groupthink, covering your heinie, etc. Of course, getting strong individual incentives into a big organization is difficult, so firms find in economic to give up some incentives to get the benefit of group knowledge in making decisions. Even in the best scenario, this tradeoff exists, so there will be some "optimal foolishness." Ideally, a robust competitive market will push firms to get to the right point of the tradeoff. Unfortunately, I fear that in the world of cronyism we are now in, the endemic problem of groupthink is worsened by the risk of offended the firm's government benefactors.

  13. Does anyone remember Thomas Hoenig, the brave lone dissenter in the November 2010 FOMC vote to activate a second round of Bernanke’s quantitative easing experiment? His opposition and dire warnings were based upon 30 years of on-the-ground Midwest banking experience, and are described well in Christopher Leonard’s book, The Lords of Easy Money.

  14. the problem to me is the idea that we must find a plan
    Instead, find ways that encourage and nurture many plans
    You make a plan better than mine, I copy it, even apes mimic
    this idea makes breadth first searches the norm
    each finds a way
    some better than others
    mimicry is waaay human
    we converge to many paths that work but are different
    not to global best but to local good enough
    After a while Global Best will be on Twitter

  15. While I sympathize with the desire to solve the world's problems, I am put off by elite attitudes in response. I am deeply skeptical of top down solutions that are sure to work perfectly as long as every one does what they are told.
    Mao used the temptation of "1000 Flowers Bloom" to lure innovation from hiding and decapitate it. Better it was adopted as a general plan, Try everything, someone may get it right.
    Liberty means making your own way.

  16. Court jesters need a well paid slot in big organizations.

  17. Too many fail to realize that all is probabilistic and therefore need to examine the certainty of their thoughts by asking : what if the proposed decision turns out to be wrong? What are the consequences and how quick can a course adjustment take place? There are no magic bullets.

  18. To understand groupthink's limitations also requires understanding its value in organizations. Call it instead a collective sense of how issues are to be considered or handled, developed over years of experience and prior thinking -- almost like a set of common law precepts. Organization members can rely on it without having to reinvent the analysis each time. Usually it's correct, or at least workable. But novel or high-stakes situations test its bounds and utility, and leaders have to be alert to those possibilities and encourage (responsible) dissenting voices and new thinking when they arise.

  19. I think the answer to the core question here is a factor hardwired into human psychology. Keeping the tribe together over rules individual thoughts because Nature required it for survival for 99% of our existence as a species.

  20. In a group you are not accountable individually and yet your individual behavior is very visible to the group. Much harder to take the risk of being the outspoken contrarian to the consensus group opinion when there is little personal downside to you. Safety in numbers.


Comments are welcome. Keep it short, polite, and on topic.

Thanks to a few abusers I am now moderating comments. I welcome thoughtful disagreement. I will block comments with insulting or abusive language. I'm also blocking totally inane comments. Try to make some sense. I am much more likely to allow critical comments if you have the honesty and courage to use your real name.