The Five Books That Could Save America

"Could," as in "if we had the capacity to govern"

I’ve long been skeptical that we, as a people, can recover from the catastrophe we now face. Most think that catastrophe is Donald Trump. I certainly believe Donald Trump is a catastrophe, but the catastrophe we face is deeper. And to cure it, if indeed it can be cured, we must work through its layers.

One layer is the corruption of money. A necessary but not sufficient condition for curing our deeper catastrophe is to end the corrupting influence of money in our politics. That’s possible, and we might be on the edge of a critical breakthrough, but even that breakthrough won’t be enough.

Because even deeper and profoundly more consequential is the corruption caused by engagement-based media. Cable news and social media render us inherently and unavoidably incoherent as a people. They profit by turning us into ignorant people who hate each other, and through this business model, they’ve earned endless profit. Solving the corruption of money won’t solve this. Given the extremism of the First Amendment, it’s not even clear we have the constitutional authority to solve this. But even if we did, there’s a third layer to our catastrophe that I’ve become convinced we must solve as well. And this layer, surprisingly (given my brand is pessimism), we could solve. At least, I think I’ve glimpsed a way.

This third layer was introduced most prominently by the first (and most famous) of the five books that could save America: Abundance, by Ezra Klein and Derek Thompson. Abundance describes powerfully and effectively the current incapacity of our system of governance. We’ve lost, as they evince and everyone notices, the ability, as a people, to do big things. The book maps a range of such failures. The common thread linking these cases is the emerging sclerosis of American governance. That, for Democrats especially, is existential, because Democrats especially proffer government to address the problems we all recognize as real.

This is a point I’ve long believed but never been able to articulate as powerfully or as effectively as Abundance does. It was the essence of my diagnosis after the last election: Democrats are obsessed with describing the 15 great geeky plans they’ve got for solving key problems in our society, never recognizing that the American public glazes over at the descriptions because the American public believes that government can’t do anything. We’ve been convinced by a line in Ronald Reagan’s first inaugural address: “Government is not the solution to our problem; government is the problem.” Anyone trying to understand this layer must start with this book.

Yet exactly one month before Abundance was published, another really extraordinary book appeared with less fanfare and attention, and undeservedly so. Because this book goes deep and broad on the thesis that Abundance articulates, and complements Abundance in important ways. Marc Dunkelman’s Why Nothing Works both gives us a big way of understanding the current moment of sclerosis and a deep way to understand exactly how profound this challenge is.

The big way is to understand the history of progressive thought as the constant alternation between two different conceptions of how government might make things better. The one conception Dunkelman calls Hamiltonian, the top-down vision of governance where a powerful governmental actor can engage to get things done. The alternative conception is Jeffersonian, which diffuses power broadly and, through that diffusion, aims to check centralized power in the name of preserving individual rights or freedoms or liberties. Progressive thought bounces between these two alternatives, often as if the two were inconsistent, as if you had to have one or the other, never both. But Dunkelman’s powerful point is that, of course, we can have and certainly need both: We need systems that can get things done, but done in a way that doesn’t violate important rights of the people or values of a nation.

The great integrity of this book is that though it argues we need to revive our Hamiltonian capacity, it begins with a powerful example of Hamiltonianism gone wrong: Robert Moses, maybe America’s last great (as in consequential) Hamiltonian. Moses had an extraordinary capacity to get things done, even the dumbest things done, even in the face of universal opposition: exhibit #1, the Cross Bronx Expressway, which laid waste to a wide range of vibrant New York culture and communities for no plausibly good reason, given the alternatives. Eventually, the public soured on this and other Hamiltonianism.

And so beginning in the 1960s, courts began enabling a practice of radical Jeffersonianism, expanding the power of citizens and interest groups to challenge the actions of government, whenever they could show that those actions affected them. Yet that rise of Jeffersonianism, Dunkelman shows, has gone catastrophically too far. And over the course of the last 60 years, we’ve replaced its Hamiltonian capacity with a system that Francis Fukuyama refers to as vetocracy. (China also accuses us of being a vetocracy, but it’s more politically correct to invoke Fukuyama than the Chinese Communist Party.)

A “vetocracy” is a system that has embedded so many veto points that it is practically impossible for anything to be done. Or more consequentially, that, because of the many veto points, it is a system that makes it trivially easy for people to extract enormous rents as the price to allow something to be done. Twenty years ago, Michael Heller called this the “tragedy of the anti-commons.” That tragedy is ours: Nothing big gets done because we’ve built a system that gives endless number of tiny forces the power to “stand athwart history and yell, Stop!”

Dunkelman’s book is extremely rich in its account of this law-driven vetocracy. The best, as in tear-your-hair-outingly frustrating, is the story of transmission lines. We will never address climate change in America until we solve the problem of transmission lines—the problem of getting clean energy to the consumers of energy, us. But just about a billion veto points block the ability to build transmission lines in America. As Dunkelman summarizes the problem:

As of late 2022, the United States boasted roughly 400,000 miles of transmission wire. According to the National Academies, to get to net zero emissions, the US will have to build 120,000 more miles of high-voltage line. In 2021, a mere 386 were added to the grid. In 2022, the lack of transmission capacity kept the delay for new projects requesting connection to roughly four years long, a duration twice as long as in 2005. And a recent study by researchers at Princeton estimated that, absent a vast acceleration of grid development, four-fifths of electrification’s environmental benefit — our growing ability to replace polluting technologies with clean alternatives — will be squandered. The problem here isn’t a lack of awareness. Nor is it that the movement’s leaders aren’t motivated to get things moving. The issue is procedural. Progressives, fearful of centralized power, have inserted so many checks that building the grid is an exercise in futility.

And so, we’re stuck. Climate change will sink us because lawyers must be lawyers, and nowhere does the system reckon the high cost of such Jeffersonian legalism.

Both books argue—powerfully, and correctly and profoundly so—that unless we can recover (in Dunkelman’s language) a Hamiltonian capacity, we are sunk. And yet the challenge of even imagining a solution is overwhelming. How can you possibly unseat a million lawyers? Who’s going to undo 60 years of judicial precedent? We’re not going to abolish federalism. We’re not going to repeal environmental protection laws. We’re not going to banish courts from challenging bureaucratic implementations of the law. We’re too far down the road building out an infrastructure for the law to make sure government does what it’s supposed to do to imagine giving that project up. These books show us the source of the problem and its extent. They beg for an innovation that might offer a solution.

Enter book 3, which I read by accident. I’m reading everything I can about AI. After finishing Sebastian Mallaby’s The Infinity Machine (highly recommended!), the recommendation algorithm in my reading world suggested I read a book called The Algorithm by Jon McNeill. The book was highly rated; I didn’t take the time to do a deep dive to understand what it was about; I just accepted the recommendation. But ironically, given an algorithm recommended it, The Algorithm is not a book about AI. The Algorithm is a management book, and worse, as I first got into the book, I thought it was a book about management at Tesla.

But I was a sucker for the story of the book because I’ve long fantasized about the life I chose not to pursue, and this book is the story of that life. My dad was an entrepreneur. When he got into some trouble, I had to leave school and come back and run his company for a period. The only thing I’ve ever done that was more fun was speaking in Iowa and in New Hampshire when I was trying to become a candidate in the 2015 Democratic primary for President (with a campaign designed to attack the first layer of this catastrophe—the corrupting influence of money in politics).

There are many days when I deeply regret not pursuing a life that would have pursued the challenge of making things work more efficiently. The Algorithm gave me a chance to live the entrepreneurial fantasy all over again. The methods it was describing were, as it promised, not actually particular to Tesla or the companies he was describing, but a more fundamental point about how innovation happens anywhere.

McNeill describes five steps of innovation—his algorithm—and the book is filled with stories about how those steps produced radically different structures for production as the people running the businesses McNeill describes (Tesla, Lululemon, GM, SpaceX, and more) learn that they didn’t actually understand their own business. And that they could restructure their work to make their business work wildly better.

The algorithm (as in the one he describes) is brilliant (or so this wanna-be innovator thinks). Seeing it as a process makes clear a pretty startling fact about how we’ve organized society.

Roughly 2/3ds of society exists in a context where many different leaders have the authority to render plastic how their organization functions, and 1/3d of society (government) exists in a context where it is practically impossible to do the same. Put differently, literally any business in America, for profit or not for profit, could apply the algorithm McNeill describes and (possibly radically) improve their work. But government can’t. Government is bound by structures that have become practically impossible to modify. Some (though if the constitution were taken seriously, not many) are constitutionally prescribed; most are the product of laws; and most frustratingly, some are the product of activist courts radically increasing the reach of those laws. The consequence is that the (in)efficiency of a large chunk of American society is frozen in place.

How could that possibly make sense? Lawyers like to speak as if the principles they enunciate are somehow glimpsed from the mind of God, as if their being fixed and unchanging makes them sensible. But the lesson of innovation and progress is that no generation knows best, and that every generation needs the opportunity to update and conform their ways to make them fit the technology of the age and the affordances that technology provides. The idea that in the 1930s the authors of the APA somehow knew best how the government should function for the next hundred years is just crazy talk.

And yet there it is: fixed and unmoving and sclerotic in its effect—unless we could imagine a strategy for updating it or for learning how best to make it work better. No one could imagine—or support (certainly I don’t) — abolishing administrative law, wiping the slate clean, and allowing administrators to do whatever the hell they want. So what, short of that, could allow the technique of The Algorithm to operate within the context of government?

The obvious answer that many have suggested in various forms is a kind of legal exemption zone. These are typically geographic areas, but not exclusively. Forty years ago, Jack Kemp pushed enterprise zones; Paul Romer has promoted charter cities; states like Nevada have created “innovation zones”; Arizona and Utah have had established “regulatory sandboxes.” Each of these is similar in form: They all draw a line around some conceptual space (geographic or jurisdictional); within the line, normal regulations don’t apply. They are either preempted or exempted, charging the administrator to achieve some objective, subject to far fewer veto-cratic constraints. (The idea is similar to the early Roman practice of appointing time/purpose-limited “dictators”: magistrates with supreme power but only for 6 months, who resigned when their purpose was accomplished. Cincinnatus is the most famous example: called from his farm to serve when Rome faced a military disaster, he defeated the enemy in a matter of days, resigned his dictatorship, and returned to farming.)

This idea applied to (what I’m going to describe as) the “abundance agenda” could work like this: Congress passes a law (yea, I know, but just assume the unimaginable for a second) that establishes project-based innovation zones. The law gives the President the power to designate a project to be within such a zone. (“If I’m elected, I commit to ….”). That designation would then give the President the power to nominate (and with Senate approval), appoint a “chief innovation officer” (CIO) for that project. That CIO would then have the power to preempt state law and exempt a project from the particulars of a specified list of federal law. That exemption would not be absolute; the CIO would need to conform to a rough equivalence of the objective of the laws from which the project is exempted. But the judgment of “rough equivalence” would be vested effectively exclusively within the CIO’s discretion. The CIO would then manage the project, spending money Congress has appropriated to achieve its objective, and designing along the way, in a The Algorithm-inspired way, a structure for ongoing regulation within the project’s domain that would be proposed upon the project’s completion for Congress to then adopt. The aim, in other words, would not simply be to build (the equivalent of) a car; it would be to build a car and design the process for building such cars going forward.

Take an obvious example: transmission lines. An innovation zone targeting transmission lines would give the President the power to appoint a transmission line CIO. That officer would then act aggressively to build out transmission line capacity, preempting conflicting state law and exempting the work from inconsistent federal regulation. Through the process of building transmission lines, the CIO would develop an understanding of a more effective process for building lines going forward. As the process is successful in achieving its purpose, it would propose the regulatory reforms necessary to return the domain to a normal regulatory regime. And if the process worked for transmission lines, imagine extending it to housing policy or high-speed rail. In context after context, we could imagine deploying the energy building big things requires, tied with a process for normalizing a more effective process within that domain.

None of this would be cheap. And if you’re to attract the right kind of innovator (McNeill types, not Musk types), the compensation would need to be significant. But depending upon the project, the benefit to all of us would be orders of magnitude greater than any cost. Again, consider transmission lines: if you believe addressing climate change to be existential, then so too is solving a necessary step to solving climate change existential.

This idea is closest to the idea that is described by Jennifer Pahlka in the fourth book that could save democracy: Recoding America. Because indeed what Pahlka describes is essentially this idea, which, in her actual practice within the Obama administration (and without the special legislative authorizations that I am describing), she effectively did. Her book is brilliantly skeptical of purely legal fixes. Instead, her objective is to describe structures that increase the implementation capacity inside government. This need she describes comes from the fact that policy-making inside government has become institutionally separated from implementation. Her proposals close that gap, if less ambitiously than the structure I am describing.

The best real-world example from her extraordinary book was the Obama administration’s response to the HealthCare.gov failure. The project, which she created and oversaw (U.S. Digital Service), rescued HealthCare.gov through a structure that insulated an implementation team temporarily from normal bureaucratic constraints. The team was small; it had direct access to decision makers; it concentrated authority in a few hands; it had the opportunity to iterate rapidly, and it could cut through the procurement and organizational barriers that blocked the ability to work efficiently. Its aim focused on outcomes rather than on process compliance. The project was not wholly exempt from law, but it operated under an exceptional governance structure—precisely the sort I’m describing, though mine is more general and potentially more effective.

The key is that it is not completely general; it is not vesting the complete authority to ignore law across the bureaucratic range (any more than McNeill was describing a process for ignoring contract or property law). It is tying that capacity to a particular project, defined within a particular scope and executed by an administrator with the talent, experience, and objective to achieve that objective, while designing a process to make achieving similar objectives within that domain more efficient going forward.

Yet there’s still one part of this story missing.

As skeptical as any sane American should be about the capacity of its government to actually execute effectively, even more fundamental is the skepticism we all have about the capacity of representative democracy to actually set policy for all of us rather than for the few best funded among us. The point, in part, relates back to the very first catastrophe that I identified we need to solve: the catastrophe of the corrupting influence of money in American politics. But it’s not just the money. Our policymakers are an embarrassment. If they’re not maximizing for campaign contributions, they’re maximizing for cable news hits or retweets on social media. Both sides practice righteous grandstanding as they try to whip up their base to defeat the other side at the next election. The world of Washington policy-making is a world of perpetual campaigning, and a world of perpetual campaigning is not a world where anything sensible gets done.

We should recognize just how fundamentally different that world is from how Washington was even just 40 years ago, before the Republican Party became competitive. In the House, Congress functioned in a relatively sensible way, relative to today, at least. Members of Congress described a regular pattern: 18 months of governing, 6 months of campaigning. During the 6 months, nothing real got done, but during the 18 months, real governing got done. We don’t live in that world anymore. We don’t live in a world where real governing gets to happen, and even where it does happen, given the catastrophe at layer two, no one is rewarded for that success. Here, the Biden administration is exhibit one. Without a doubt, Joe Biden was the most effective president since LBJ, shepherding hugely significant legislation through Congress in a bipartisan way. That legislation was profoundly important. Yet despite this success of truly historic character, popular support for President Biden was never significantly better than the popular support for Donald Trump in either his first or second term. Doing good (for the country) does not translate into doing well (politically). That truth will be an important lesson to presidents going forward.

So how could we solve this problem of policy choice? How could we get our government to make policy decisions based on what makes sense for all of us rather than what makes sense for the richest or loudest or most skilled vetocrats?

Enter the final of these five books that could save America, Hélène Landemore’s Politics Without Politicians. Landemore is a Yale political scientist who has walked her talk. She has long theorized about the core innovation that democracy needs (such as in her first book, Open Democracy (2020)). But since that book, she has been deeply engaged in the actual practice of the innovation that could save sensible policy making in America and elsewhere: the citizen assembly.

A citizen assembly is a large body of randomly selected and representative citizens, gathered for a particular purpose, provided material to help them understand the issue they must address, and given the chance to deliberate on the issue for as long as the issue requires. (Think a jury, but actually representative.) Stanford professor James Fishkin has advanced a related, if more formal practice — the deliberative poll — for the past 30 years. Versions of the idea have been part of self-government across the globe since ancient Greece.

And as Landemore describes in a book destined to become the Bible for the modern citizen assembly movement, the idea has been revived across the world, as confidence in elective democracy has collapsed everywhere. Across the world, citizen assemblies are demonstrating that ordinary citizens, if representative, can be brought to understand and deliberate upon complex questions of policy and, from that deliberation, produce policy judgments that are, critically, impressively sensible.

A few in this movement carry the idea to its logical end — lottocracy, which replaces elective representatives with randomly selected citizens. Landemore does not go that far, at least not yet. In her conception, empaneled bodies of representatives complement a policy-making process, providing democratic credibility to a recommendation grounded in an informed and deliberated democratic procedure.

This is the complement the abundance agenda needs. In the early 20th century, the alternative to politics was administrative agencies governed by experts. The world has grown weary of experts as it has grown skeptical of the elite generally. But everywhere, citizen assemblies are demonstrating a capacity to understand and determine even complicated questions effectively. In Texas, for example, almost 30 years ago, citizen empaneled deliberative polls have resolved complex questions of energy policy — leading that oil-rich state to become a leader in clean energy as well. And in France, a citizen assembly composed of more than 150 citizens from across the nation spent 9 months deliberating on climate change proposals, concluding with 149 proposals for reducing France’s greenhouse gas emissions.

The abundance agenda should embrace the citizen assembly movement. When a difficult policy question must be resolved, the CIO could convene an assembly, and the assembly could resolve it. Critically, the process would require informed and extensive support, both to ensure participants understand the issue and have a chance to deliberate upon it fully, but also to ensure that a genuinely representative group deliberates. Done well, the assembly could add democratic credibility to difficult choices of policy, giving the CIO a direction to implement.

Drawn together then, the lesson of these five books is this: Klein, Thompson, and Dunkelman describe the challenge of the abundance agenda — the critical need for democratic governance that once again could govern. McNeill and Pahlka give us a glimpse of a practice of innovation that could make government effective again. The challenge to such a practice, however, is the risk of undemocratic policy choice, either fixed on experts or autocrats. Landemore gives us a way around that problem, vesting these choices in an informed and deliberated representative public. Together, the five identify the problem that democracy now faces and suggest a critically democratic way to solve that problem. Together, they are the hope elective democracy needs.

Don’t get me wrong: I recognize just how impossible all this sounds. It is a classical fallacy of reform writing to describe a problem and then a solution that ignores the problem just described. That’s precisely what I’ve done here. I’ve not explained how the corruption of money gets addressed; I’ve not even argued we have the constitutional authority to fix the corruption of engagement based media. The vetocracy will do everything possible to veto this reform. Any idea linked to Tesla is inherently tied to Musk, which is unavoidably linked to the greatest failure in bureaucratic reform in modern times: DOGE. And finally, for most, the idea of a “random” group of citizens doing anything sensible just sounds oxymoronic.

These challenges are real and likely fatal. But my point is not prediction. I’m not saying the idea these five works suggest is likely. I’m saying it is necessary. Whether and how we escape the constraints that make this all but impossible is a question for others. It is enough—it is everything—to see a way forward.

Because if we don’t find a way forward, this republic is at an end. We are being defeated by climate change. AI presses an unprecedented transformation in the economic life of the vast majority of Americans, not to mention the runaway risk most of those most knowledgeable about the technology fear most. We literally could not face a greater need to rediscover a capacity to govern ourselves.

Just as the alcoholic, as he loses his family, his job, his friends, his health, could not face a greater need to stop drinking. We all know that’s not motivation enough for most severe alcoholics. Most are defeated by their disease.

And so too is it likely for us.