Down With the System, The System's Not Down
The Fifth Risk and The Revolt of the Public and the Crisis of Authority in the New Millennium.
“You might have good reason to pray for a tornado, whether it comes in the shape of swirling winds, or politician. You imagine the thing doing the damage that you would like to see done and no more. It's what you fail to imagine that kills you.”
-Michael Lewis, The Fifth Risk
I remember exactly where I was on June 23rd, 2016. I sat agape at dinner, with an irritated girlfriend and a half-finished bucket of crab, ostensibly enjoying a brief beach vacation, staring at my phone as I repeatedly refreshed the BBC website. The Brexit votes were being tallied. Polling heading in to the vote showed a very close race, though most predicted the British would choose to remain in the EU. I knew the EU had its issues - it had only been a few years since the daily headlines about the viability of the Eurozone and an economically crippled Greece. But I also believed that rising global trade and a worldwide web were knitting the world more tightly together and that international institutions like the EU would grow more powerful as a result. Wouldn’t politics follow economics and culture? With those priors, it was hard for me to believe that the UK might really decide to leave, despite the warning signs. I might have avoided my failure of imagination had I read Martin Gurri’s The Revolt of the Public and the Crisis of Authority in the New Millennium when it was published two years earlier in 2014.
The Revolt of the Public is a sweeping thesis on the impact of freely available information on nearly every sphere of human life by former CIA analyst Martin Gurri. Gurri argues that public access to information, supercharged by the internet and social media, has increased the influence of individuals and hamstrung institutions from governments to universities to corporations (he uses the term “institution” very broadly). Unable to adapt to the increasing pace of change or to hide their failings from the public, institutions are losing power and authority, with nothing (as yet) to replace them.
Institutions were never more powerful than in the 20th century. New information technologies like the radio and television broadcast information from one to many, with a few key publishers and media companies serving as gatekeepers for the information diet of the public. Institutions had the resources and access to control the narrative of their relationship to the public. They highlighted their successes and buried their failures. This monopoly on information made institutions persuasive; it seemed like the CEO in the boardroom, the politician in the legislative chamber, and the professor at the lectern had matters under control. Yet change was coming.
Gurri believes the explosion of available information via both digital and traditional media in the 21st century negates the ability of a wide array of institutions to control their narratives. Information today can come from anyone and go viral. Without control, institutional failures (e.g. a bad policy or a failed forecast) are shared widely and erode the lifeblood of an institution: trust, legitimacy, and authority. For Gurri, the framework of “The Center vs. the Border,” created by Mary Douglas and Aaron Wildavsky, can explain much of the interactions between institutions (the Center) and the public (the Border):
“The Center, Douglas and Wildavsky write, is dominated by large, hierarchical organizations. It frankly believes in sacrificing the few for the good of the whole. It is smug about its rigid procedures. It is too slow, too blind to new information. It will not believe in new dangers and will often be taken by surprise (…) The Border, in contrast, is composed of “sects”—we would say “networks”—which are voluntary associations of equals. Sects exist to oppose the Center: they stand firmly against. They have, however, “no intention of governing,” and develop “no capacity for exercising power.” Rank means inequality, hierarchy means conspiracy to the Border. Rather than articulate programs as alternatives to those of the Center, sects aim to model the behaviors demanded from the “godly or good society.”
Having set the stage, Gurri transitions to highlighting examples of conflicts between publics and authorities across a wide range of public life. His analysis is most compelling when focused on politics, with examples from the Arab Spring to the Indignados in Spain, to Occupy Wall Street in the US. In each case public uprisings, many led by well-educated and wealthy members of the upper-middle class, laid blame for a range of societal evils at the feet of incumbent institutions and demanded sweeping change. Gurri doesn’t stop with governments, however. He extends his analysis to corporations and to scientists and highlights loss of public trust in these authorities as well, though frankly, his examples are weaker.
Authority requires legitimacy. In the pursuit of legitimacy Gurri thinks institutions now just want to make their constituents happy. In so doing, they take on an ever-increasing array of problems:
“From obesity to climate change, nothing is so personal or so cosmic that it can’t be reckoned a failure of government. If political power has become the guarantor of happiness, then politicians must take the blame for the tragic dimension of human life.”
As governments expand their responsibilities, they create more opportunities for failure. Gurri notes institutions fail all the time and argues that systemic and societal problems are too complex to be solved by big pieces of legislation or regulation. Change happens too rapidly, and institutions lack the nimble adaptability required to respond to the fast-moving networks of The Border. This in turn, erodes trust and jades the public, which then attacks the institutions, starting the cycle all over again. Gurri’s ultimate fear is that the public will lose faith in democracy itself (a fear which seems well-founded given the events of the last year).
Gurri’s book outlined a useful mental model when the book was initially published in 2014 – and has proven a handy guide to a variety of events since then. While Gurri didn’t predict Brexit, Trump, the US response to Covid-19, or GameStop and AMC, he provides the tools to analyze the relevant actors (publics and institutions) and their intentions. The trends outlined in the book have continued apace - since the publish date, trust in all information sources has fallen to varying degrees, according to the Edelman Trust Barometer. Just look at how the annual titles to the survey have changed since 2000!
Trust plummets after the Great Recession, and never really recovers, with the back half of the 2010s a bleak march of “trust inequality”, “trust crises”, and “trust battles” before our apparent collective “declaration of information bankruptcy” this year. Perhaps, as in bankruptcy, we get a chance to start fresh. I suspect Gurri isn’t holding out much hope.
Gurri does have some solutions – for us, as members of the public, and for governments and other institutions. The public ought to lower their expectations of institutions, which will continue to fail frequently, and to pick honest, humble leaders who can breathe new legitimacy into old institutions. Institutions also need to lower their sights, to incorporate the public into a positive vision of the future, rather than a dystopia, and to acknowledge their imperfect knowledge. Gurri’s last line is “The reformers of democracy must learn to say, out loud for all to hear, “This is a process of trial and error,” and, “We are uncertain of the consequences,” and even, “I was wrong”.
Gurri is perhaps too pessimistic about the potential for the public to build new institutions. Over the last 10 years, large communities have worked on building decentralized institutions of governance in a variety of fields, from remote work, to charter cities, to digital cryptocurrencies. But this is not the part of the essay where I extoll the benefits of blockchain or other experiments in distributed and trustless networks. Instead: a defense of governmental and institutional competence when facing problems of immense complexity.
Michael Lewis’s 2017 book The Fifth Risk was the scariest thing I read that year. In three parts examining the Departments of Energy, Agriculture, and Commerce, respectively, Lewis reported how essential but little-understood parts of the US government were (mis)managing the transition to the Trump Administration. But a re-reading in 2021 after Trump left office offers a different interpretation: how effective governments and institutions can be in adapting to and managing complex systems.
The Department of Energy (DOE) shows governments have been successfully managing complex systems with a very high cost of failure for decades. Half the DOE budget ($15B) goes to managing the US nuclear weapons stockpile. The institution has to be robust to all types of unexpected incidents, from “Broken Arrows,” or 32 global nuclear weapons accidents since 1950, to the more mundane mix-ups:
“The list of things that might go wrong inside the DOF was endless. The driver of a heavily armed unit assigned to move plutonium around the country was pulled over, on the job, for drunk driving. An eighty-two-year-old nun cut through the perimeter fence of a facility in Tennessee that housed weapons-grade nuclear material. A medical facility ordered a speck of plutonium for research, and weapons-lab clerk misplaced a decimal point and FedExed the researchers a chunk of the stuff so big it should have been under armed guard-whereupon horrified medical researchers tried to FedEx it back.”
The DOE has its issues – somewhat famously, there was so much workforce turnover in the 2000s that the agency forgot how to make a key classified raw material for nuclear weapons and had to reverse-engineer its manufacturing. Yet even with four years of Trump, who showed limited interest in the DOE (he appointed Rick Perry to lead the department only a few years after Perry included the DOE in the list of agencies he said he’d eliminate entirely if he were elected President), there have been no catastrophic accidents. Luck plays a role, but so does institutional resilience.
Government institutions can do more than play defense against complexity and tail risk. They also act and adapt to complex situations with as much savvy as the best VC or tech company. Gurri highlights the importance of trial and error, and making small asymmetric bets with high potential rewards. The DOE has a portfolio of low-interest loans to encourage innovation in alternative energy to make the big names on Sand Hill envious.The program, called ARPA-E provided the initial financing for the Tesla factory in Fremont, CA and was one of the earliest backers for the emerging utility-scale solar industry, back when no VC was willing to take the risk.
Government agencies are also on the cutting edge of using data to better tame and predict complex systems – precisely the kind of project Gurri believes is too ambitious. The National Weather Service, housed in the Department of Commerce, has a multi-billion-dollar data gathering operation (they collect twice as much data as the entire book collection of the Library of Congress on a daily basis) to predict the weather, one of the most complex systems we have. And it’s working. There has been an “astonishing improvement in all weather predictions. The five-day-out forecast in 2016 was as accurate as the one-day-out forecast had been in 2005. In just the last few years, for the first time in history, a meteorologist's forecast of how hot it will be nine days from now is better than just guessing.” Predicting the weather has saved, and will save, countless lives – 8,000 people died in a Category 4 hurricane in Galveston, Texas in 1900, while the last five Category 5 hurricanes to hit the US (Andrew, Dorian, Michael) caused a combined 43 fatalities. This achievement appears to be underappreciated – there are no histories of the National Weather Service on Amazon or Google Scholar, and the only history of its parent, the National Oceanic and Atmospheric Agency (NOAA) appears to be an internal memo published in 1987.
On communication, Lewis and Gurri agree – the government (and perhaps institutions more broadly) don’t do a great job of interacting with the public and balancing out the narrative of failure. Lewis highlights that “the United States government had a better handle on the weather than on its own people. It had spent billions of dollars to collect data about the weather, and none about how people responded to it.”
Gurri points out how lowly US government webpages ranked on Google in 2014 – and this is a problem which certainly has not been solved. The DOE, a critical agency managing resources which, if misused, could wipe humanity off the face of the planet, has averaged about as much interest on Google as the second most popular NBA team in Los Angeles over the last five years:
Lewis spends a lot of time in the book highlighting the importance of Presidential transitions, as new administrations are often uninformed about the role of a particular department or agency in the function of society. Similarly, perhaps the government should spend some more time on effectively educating the populace too – hiring the Wendy’s Twitter guy, for instance, or providing meme training (I’m serious). There may be a greater appreciation of the institution we have as a result, and what we have at stake.
Read/Skim/Skip: Here’s my guide for how to approach both books, should you choose to read further.
The Revolt of the Public and the Crisis of Authority in the New Millennium: The forward to Gurri’s book contains a recommendation – read the last chapter, titled “Trump, Brexit, and Farewell to All That” twice. It was added when Stripe Press re-published the book in 2018. The forward recommends reading the chapter before the rest of the book, and after. I did this, and found it helpful, though frankly the first read-through should probably be a skim.
Read: Chapter 1: Prelude To A Turbulent Age and Chapter 3: My Thesis, which together outline the main components of Gurri’s Thesis. A very quick reading can focus on just these two chapters and Chapter 9: Choices and Systems, where Gurri outlines his solutions to the problems he outlines. A more in-depth reading should also include Chapter 8: Nihilism and Democracy, where Gurri goes more in-depth on the impact of his thesis on democratic politics.
Skim: Chapter 2: Hoder and Wael Ghonim, Chapter 5: Phase Change 2011, Chapter 6: A Crisis of Authority, Chapter 7: The Failure of Government. These chapters are full of the examples Gurri uses to support his core thesis, and you can skim through the details and focus on the examples you find most interesting. I think Gurri is somewhat overly focused on recent examples at the time he was writing the book – for example President Obama gets a lot more treatment here than he might have had this book been written in 2021, or 2040.
Skip: Chapter 4: What the Public is Not. If you’re feeling argumentative or skeptical about how he defines his terms, maybe come back and look for a counterargument or response to your skepticism here – otherwise, you can skip. The same goes for Chapter 10: Finale for Skeptics. Gurri outlines a null hypothesis and argues his core thesis disproves it.
The Fifth Risk: Read it all. The book is relatively short at 219 pages, and a quick read organized into three parts. I touch on two of the three above, and leave out the chapter on the Department of Agriculture, but for those interested enough in the topic to buy the book, it’s worth a read as well.
Thanks to my On Deck Writer’s cohort (Kate Long, Harry Moy, and Matias Sanchez Sarmiento in particular) for feedback on drafts of this piece. If you end up reading either book, feel free to hit me up on Twitter at @kyleallenniesen. I’d love to know your thoughts. And if you happen to be in LA, let’s meet up - coffee is on me!
Gurri points out that the lifespan of an average company listed on the S&P 500 has declined from 67 years in the 1920s to 15 years in 2014. and argues corporations today struggle to keep up with the rising pace of innovation. This ignores significant changes in the capital markets landscape (the rise of private equity, which didn’t exist in the 1920s) and an increase in mergers and acquisitions. Michael Mauboussin highlights in a 2020 paper that a newly listed public company in the 2000s had a 63% chance of surviving five years, down from a 92% chance for companies listed before 1970. He draws nearly the opposite conclusion to Gurri, noting that most firms de-list due to acquisition, and more and more industries are dominated by a small number of superstar firms, as measured by the Herfindahl-Hirshman Index. Corporations are in some ways more powerful than ever today, which runs counter to Gurri’s core thesis.
ARPA-E, has, as of their last annual report for FY 2018, dispersed $1.7B in grants. Project teams had attracted $2.6B in follow-on private investment, which was inflecting rapidly - $1.4.B of that follow-on came in just the previous two years.