World Issues

Prospects for a Multi-Polar World

Dan Scott

15 Jan 2023 • 37 min read

Contrasting and synthesizing the perspectives of Balaji Srinivasan and Peter Zeihan. Skim down to the section on "Contrasting Geopolitical Outlooks" if you want to get to the main part of this post more quickly.

Introduction

The present moment certainly feels like we can't take anything for granted anymore. Recent events such as the on-going war in Ukraine and the continuing after-effects of the Covid-19 pandemic have put a lot of strain on the established geopolitical system. Maybe everything will return to normal after a bit more time to adjust. But what if it doesn't? In fact, even before the past few years, many observers were questioning how much longer the geopolitical status quo had in it (e.g. whether and when an alternative global reserve currency to the US dollar would emerge).

I recently read a couple of books that touch on this question of "What's Next?". The Network State by Balaji Srinivasan is a meta-manifesto for communities rooted in the cloud to fill the middle ground between anarchy and tyranny. The End of the World is Just the Beginning by Peter Zeihan is a guide to the strategic geography of a deglobalized world. Both books were released last year (2022). They have very different views on how the relative strength of the USA and China will trend over the next decade or two, although both expect major changes from the status quo. From the outset of this meandering review of both books, I want to emphasize the value in reading the two of them. If you read a book with punditry and prognostications in isolation, it's easy to think the author has it all figured out and be too credulous about their conclusions; reading multiple authors with different views is a way to uncover their assumptions and dig into the strengths and weaknesses of their arguments. I have a lot of respect for the intelligence of both of these men, but they aren't prophets. So this post isn't about figuring out what will happen geopolitically in the 2020s, but rather to try to conceptualize the possibility space. And for that task, points of disagreement can be as illuminating as points of agreement.

Each author brings a unique focus and, conversely, various blindspots of their own. Balaji Srinivasan places a lot of emphasis on how technology interacts with and enables political possibilities. He's also big on the importance of healthy institutions. Peter Zeihan's focus is more on the way geography (access to natural resources, ease of internal transportation, and defensible borders, to list some key factors) determines national outcomes. Along with this, he puts a lot of weight on how demographics—notably the age pyramid shape—inexorably influences economic trends (Hans Rosling also emphasized the reliability of demographic projections). Understanding their different analytical frameworks can shed light on some of the places their predictions differ. For example, Balaji's focus on technology means that he doesn't think a large population of working/fighting-age adults is as important a factor in the strength of a country anymore because he thinks automation will be able to take up a lot of slack (e.g. industrial robots and combat drones).

Another crucial piece of context for understanding their work and forecasts is where each author lives. Balaji Srinivasan was born and raised in the USA, but recently moved to Asia (Singapore, with considerable time also spent in India, from what I understand). This reflects his view that competant governance and national unity is important. Peter Zeihan lives in the Rocky Mountains near Denver: a defensible location adjacent to the most important agricultural region in the world. This reflects his view of the importance of reliable transportation (river or rail) of vital natural resources (food, energy, and materials). To be fair, I don't actually know why each author lives where he does, but they are both thorough and careful enough thinkers that I doubt it's entirely coincidence that their choices of residence reflect their outlooks.

(Before I get further into these reviews, a quick note about names: I'll mainly refer to Balaji Srinivasan by his first name and Peter Zeihan by his last name. This isn't a sign of any difference in the level of respect or familiarity towards these authors but simply follows from how they most commonly refer to themselves. See for example, their website URLs or main social media accounts.)

In the following sections, I'll provide an overview of each book, get into the core of their disagreements, then cover less central material from the books and related articles for further reading.

The Network State

The theme of this post is not actually the main theme of The Network State. It is about, well, Network States. What are those? The front matter gives some useful summaries of various levels of detail, such as this sentence:

A network state is a highly aligned online community with a capacity for collective action that crowdfunds territory around the world and eventually gains diplomatic recognition from pre-existing states.

The geopolitical outlook conveyed in this book comes up largely in the context of why the time is right for this type of governance innovation. I'll get to that in a subsequent section; in this section I wanted to cover some of the other material from the book that is worth discussing.

One of the things that stands out about The Network State is its structure. Right at the beginning, it has a Quickstart section that provides executive summaries of various lengths: one sentence, a 7-point outline, and an essay. More books should do this. Each chapter also begins with a pretty clear overview of what it is going to cover, with more detail added as you read on. This gives the whole book a bit of a fractal feel: repetition at increasingly fine-grained scales. So a totally valid way to read it would be to check out the Quickstart section and then drill into the parts that interest you the most. I should also note that it is written as a native e-book, and makes a lot of use of the features that format enables (e.g. hyperlinking) instead of being like a traditional book that you're just happening to read on a screen; it is also freely available online in its entirety.

Balaji Srinivasan is an extremely-online venture capitalist, and this really comes out in the book he's written. For one thing, he has an obvious hobby-horse about ethics in tech journalism, and instances of the New York Times being wrong are collected with the passion of a philatelist and shared in footnotes or bullet points. Furthermore, he really brings what I think of as a Silicon Valley mindset to the topics he discusses (this is not a bad thing in my view—this mindset has led to breakthroughs in solving problems in other domains, so it is at least worth trying in politics): being data-driven (e.g. setting up dashboards with key metrics), focusing on founders/start-ups instead of trying to solve new problems in existing institutions, rapidly iterating instead of getting bogged down in the planning/approvals phase ("move fast and break things"), and hunting for "unicorns" (venture capitalists can accept most of their investments failing because the few that win big have such enormous upside that it makes up for a lot of losses).

The Network State has the feel of a political manifesto but I'd describe it as more of a meta-manifesto. Balaji doesn't tell readers what kind of network state they should start (although he does offer a few example suggestions); rather, he lays out a roadmap for how people with various ideas for intentional affinity-based polities could attempt to put them into practice. He also covers the general case for why someone might want to do so, and some reasons why the time might be ripe; these themes build on grand theories of history and politics that he presents.

The roadmap Balaji presents contains some genuine innovations over the conventional how-to-start-your-own-country genre. First of all, instead of starting with some particular territory (whether a corner of unclaimed land due to border anomalies, a platform built in international waters, or a ZEDE), he recommends starting by building a community (around some moral or lifestyle differentiator) and acquiring land afterwards. A network becoming a nation, as he puts it. The other twist is the weight he places on seeking diplomatic recognition instead of merely de facto sovereignty. This is from hard-learned lessons from Bitcoin, that you need some legitimate interface with existing systems. The initial community is built in the cloud and should have a distinctive moral innovation (what he calls a One Commandment)—you're not likely to succeed with a "Year Zero", written-from-scratch society; it's better to "fork" existing society on one point of disagreement/innovation (you can add more reforms later, iteratively) where you can make a moral argument for a better way of doing things. Because the community starts in the cloud, it doesn't need all of its members to be in one place:

Unlike an ideologically disaligned and geographically centralized legacy state, which packs millions of disputants in one place, a network state is ideologically aligned but geographically decentralized.

Technological developments have enabled this kind of geographically dispersed type of community:

Social networks change many things, but a critical one is that they change the nature of community. Your community is your social network, not necessarily the people who live near you. When the network identity is more salient than the neighbor relationship, it challenges the very premise of the Westphalian state, which is that (a) people who live geographically near each other share values and (b) therefore laws should be based on geographic boundaries. The alternative is that only people who are geodesically near each other in the social network share values, and therefore the laws that govern them should be based on network boundaries.

What makes network states an attractive innovation (at least to some people)? One answer is an increased ability to "vote with your feet" (inspired by Hirschman's concept of voice and exit):

migration is as powerful a way to change the law under which you live as election.

Additionally, a new frontier has a lot of value for promoting/allowing innovation and serving as a relief valve for domestic political polarization:

Everything changes when the frontier opens up, when there is a new realm of unoccupied space, where resources are suddenly less scarce. There’s less obligate wrangling, because an aggrieved faction can choose fight or flight, voice or exit.
...
So while the frontier is not a panacea, it is at least a pressure valve. That’s why reopening the frontier may be the most important meta-political thing we can do to reduce political conflict.

And the digital world is a logical place to root a new frontier:

Today, if we assess where we’re at, there are four possibilities for the frontier: the land, the internet, the sea, and space. Right now, there are 7.7B people on land, 3.2B on the internet, about 2-3M on the high seas, and less than 10 currently in space.

Finally, facilitating a huge expansion in the number of polities would ground political and social theories in more empirical data, which existing countries could learn lessons from about what works and what doesn't:

network states allow you to run ethical experiments on policy, with opt-in participants that are as interested in governance innovation as you are.

Similarly, Balaji discusses how cryptocurrencies have provided a surge in the amount of experimental data in macroeconomics. He clearly states his goal/wish that network states could make history and political science more experimental disciplines. And in The Network State he shares some theories of history and politics that he sees as decent starting points for developping testable models.

The following excerpts give a sense of the quantitative approach he'd like to bring to history:

If you’re technically inclined, you might why we spend so much time on history in this book. One answer is that histories are trajectories of dynamical systems. If you can spend your entire life studying wave equations, diffusion equations, time series, or the Navier-Stokes equations — and you can — you can do the same for the dynamics of people. In more detail, we know from physics (and Stephen Wolfram!) that very simple rules can produce incredibly complicated trajectories of dynamical systems.

From a statistician’s perspective, history is necessary for accurately computing the future. See any time series analysis or machine learning paper — or the Kalman filter, which makes this concept very explicit. To paraphrase Orwell, without a quantitatively accurate record of the past you cannot control the future, in the sense that your control theory literally won’t work.

Censorship and misinformation are both seen as threats to having reliable historical data to work with. This focus on a data-driven approach is part of what I meant earlier about an obvious Silicon Valley perspective on the part of the author. He also takes the view that witnessing the way start-up companies morph their culture as they become bigger and more established provides some transferrable insights into social and bureaucratic dynamics in general:

Most people haven’t studied enough history to have an intuition for cyclicity on a 100-year or longer timescale. But many people are familiar with the lifecycle of successful tech startups, which exhibit this behavior on a 10-year timescale. That’s about the longest kind of experiment we can run repeatedly within a human lifetime.

A willingness to experiment and then pragmatically shift course as needed is also mentioned as a strength that tech folks can bring to the challenges described in this book:

Because causality exists. Because we can run controlled experiments. Because human action can influence outcomes. Because we aren’t communists that believe in the historical inevitability of utopian outcomes, but technologists that believe in individual initiative subject to practical constraints.

I'll get into the geopolitical forecasts Balaji makes in another section below, but as a high-level summary, he expects increasing social fragmentation and political dysfunction/division in the United States; on the other side of the world in China he expects digital totalitarianism to grow in scope and effectiveness. The rest of the world will be torn between these models, while cryptocurrency and related technologies will make an anarchic "none of the above" more of a viable option in both domestic and international affairs, making it harder for any kind of moderate centre to hold. This anticipated political environment presents both a need and an opening for the type of alternative an idea like network states can offer.

The educational background that Balaji brings to his ideas and writings gives him a preference for theories/models of history and politics that make testable (ideally quantitative) predictions. For example, he refers to the cyclical models of Peter Turchin and Ray Dalio which predict that the coming decade is on schedule for conflict and a likely shift in global hegemony. He also offers a thesis that is kind of technologically deterministic, where the prevailing technologies a generation or two ago favoured centralization, while the ones at the forefront now favour decentralization; the outcome of this model is a future (in a political sense) that looks more like the US pre-New Deal. In domestic politics, he has a model where the political axis/spectrum swings around over time so that Left and Right have different positions on the object level from decade to decade, but represent stable coalitions. In fact, he argues that tribal alignments are often more stable than the ideologies they tout at any given time (e.g. VA and MA voting differently, US and Russia in opposition whether over Communism or Irredentism).

The fourth order approach is to then note that this (real!) axis actually rotates over time. It’s more about relative tribal positioning (voting with members of the same political tribe) than absolute ideological positioning (voting for a constant ideological position).

Balaji is heavily focused on technology, and his views on that inform some of the material in his book. One relevant tech trend is artificial intelligence. Not only could it help with automation (allowing productivity growth even with stagnant or declining working-age populations), it could also be a boon for authoritarian surveillance and/or facilitate the even more rapid production and dissemination of misinformation:

the advent of AI allows highly realistic fakes of the past and present to be generated. If we’re not careful, we could drown in fake data.

I could see this turning into a significant arms race, and it's not clear to me whether the ability to monitor and analyze enormous amounts of data or the ability to generate digital "chaff" will advance more quickly.

Another relevant tech trend to The Network State is the concept of Augmented Reality (AR). Requiring the right credentials or digital passport to even "see" certain things overlaid on the real world would provide some value-add that membership in a Network State could offer.

even more of our daily experience will blend not just the physical world dominated by natural law, but the digital world run by human-written code. The offline world still exists, physics and biology still exist, but algorithms and databases run even more of human existence.

These views on technology are a notable point of contrast with Peter Zeihan. As you'll see in the next section, Zeihan doesn't think any new technologies will fundamentally transform the old strategic necessities of a healthy population structure, access to vital natural resources, and defensible borders.

I'll skip a lot of the specific roadmap to starting a Network State for the purposes of this review, but if you're interested in that or in exploring some of the ideas I did cover in this section in more detail, you can read the book or listen to this podcast interview with the author.

Finally, I'll note in advance of the sections below that Balaji's predictions (for the US and China, etc.) are different from his values. His moral vision for network states is expressed as follows:

A better answer might be: none of the above. That instead of choosing either anarchic decentralization or coercive centralization, we choose volitional recentralization.

The End of the World is Just the Beginning

Peter Zeihan is extremely bullish on the United States. Or at least in comparison to the rest of the world. He doesn't deny the political division and dysfunction (as he says in a quote I excerpt below, "the Americans are, like, really bad at government"), but thinks that the natural advantages of the US outweigh those of almost any competitor. The first major advantage he sees is geographical: the core of a continent full of natural resources, with defensible borders and easy internal transportation. The second major advantage is demographic: compared to other first world countries (aside from France, and a few small players like New Zealand), birth-rates in the US have stayed high enough to avoid the worst effects of an aging and shrinking population. That is something that he sees wreaking havoc with other developed countries through the 2020s and 2030s. Japan has been able to mostly cope with an aging and shrinking population (although their economy basically stopped growing in the mid-1990s), but few other countries have their advantages (a very unified and cohesive culture, and starting to prepare and shift their economy decades in advance of crisis hitting). The current strain on Canada's healthcare system is very much in line with his thesis: a wave of retirements from baby boomers with insufficient new workers trained to replace them will be a phenomenon affecting most of the developed world for the foreseeable future. At the same time as this demographic contraction and all the challenges that come with it (labour shortages, liquidity or solvency problems for pension funds and social services), countries will have to adapt to a retreat of globalization. The US has provided the principal security guarantees for international trade (Zeihan explains that this is not just directly from allowing ships to safely cross the ocean but indirectly because countries that don't fear invasion are more willing to specialize instead of insisting that domestic production needs to focus on strategic essentials) but is turning in more of an isolationist direction as they conclude that the global order that made sense during the Cold War doesn't meet their current needs as well. With more of a sense that regional power blocs are up for grabs, Zeihan expects increasing global conflict and decreasing international trade as countries take efforts to shorten and secure their supply chains. Consider how cutting edge computer chips are only made in Taiwan. Now this is seen as a vulnerability in case conflict breaks out there, so other countries are planning for domestic production. This is a duplication of effort that leaves everyone less prosperous under normal circumstances, but lessens the risk of supply chain disruptions leading to shortages. Repeat this story across numerous industries. Importers of basics like food or energy are likely in for a very bad time if his predictions pan out.

Let's flesh some of these points out a bit. To start with, here is part of the—cheery!—introduction that Zeihan provides:

During the past seven decades, as a percent of the population, fewer people have died in fewer wars and fewer occupations and fewer famines and fewer disease outbreaks than since the dawn of recorded history. Historically speaking, we live in an embarrassment of riches and peace. All of these evolutions and more are tightly interwoven. Inseparable. But there is a simple fact that is often overlooked. They are artificial. We have been living in a perfect moment. And it is passing. The world of the past few decades has been the best it will ever be in our lifetime. Instead of cheap and better and faster, we’re rapidly transitioning into a world that’s pricier and worse and slower. Because the world—our world—is breaking apart.

The title of The End of the World is Just the Beginning refers not to the literal end of the world but to the unravelling or fracturing of the international Order (with a capital O; take Bretton Woods as a metonym for the whole system of American hegemony for a sense of what Zeihan means be the Order) that has prevailed in the West since WW2 and in most of the world (aside from very marginal places like North Korea) since the end of the Cold War. He argues that the Order is on its last legs because the conditions under which it was developped no longer obtain.

Let’s begin with base structure: part of why American manufacturers feel cheated by globalization is because that was the plan. The core precept of the Order is that the United States would sacrifice economic dynamism in order to achieve security control. The American market was supposed to be sacrificed. The American worker was supposed to be sacrificed. American companies were supposed to be sacrificed. Thus anything that the United States still manufactures is a product set for which the American market, worker, and corporate structure are hypercompetitive.

Most people think of the Bretton Woods system as a sort of Pax Americana. The American Century, if you will. But that’s simply not the case. The entire concept of the Order is that the United States disadvantages itself economically in order to purchase the loyalty of a global alliance. That is what globalization is. The past several decades haven’t been an American Century. They’ve been an American sacrifice. Which is over.

Assuming there is a roll-back of globalization, Zeihan draws a straight line from there to severe shortages of basic commodities (up to and including famine, for places that suddenly can't import either enough food or enough fertilizer and energy to grow it themselves) in places that aren't blessed with good access to natural resources. Much of the book goes over specific industries and resources, and how their distribution might change if supply chains are forced to get a lot shorter.

Here are the keys to success in Zeihan's model:

geographies of success = territory with a lot of natural resources (including good agricultural land and even things like strong wind and solar potential), easy internal transportation, and natural/defensible borders
favourable demographics = enough young people to keep the workforce strong, and not too high a dependency ratio that consumes a lot of social spending
enough naval force projection to secure trade routes to vital natural resources a country lacks domestically
neighbours with economic synergies, namely labour forces on different rungs of the skill/cost ladder so that comparative advantage can work on a regional basis instead of a global basis

The Americas (both North and South) look pretty good on these points (meeting 2 or 3 out of 4), and so do some select other countries: France, Australia, Japan, Sweden, the UK. On the other hand, China, Korea, and Germany are probably the most reliant on the continuation of globalization as they are export-driven economies that rely on imports for a lot of raw materials and necessities.

The End of the World is Just the Beginning has effective explanations on demographic factors and how they are basically locked in at this point (the right time to fix a declining birth rate is twenty years ago...):

Demographics tells us that the number and collective volume of mass-consumption-driven economies has already peaked. In 2019 the Earth for the first time in history had more people aged sixty-five and over than five and under. By 2030 there will be twice as many retirees, in relative terms.

This generates a double hit to labor markets. The Baby Boomers are the largest-ever generation, so their absence is hugely impactful in numerical terms. They are also the oldest economically active generation, meaning that their numbers comprise the bulk of all available skilled labor. Remove so many high-skilled workers in a short period of time and labor shortages and labor inflation are a foregone conclusion for years to come.

"Retirees no longer have new income to invest," so expect cost of capital (e.g. interest rates on mortgages as a very timely and relevant example) to be higher than the 2010s for a long time. It's not just a shrinking workforce to deal with as the population ages into retirement, but a whole cohort of investors switching from putting money into the market to taking it out. Additionally, older consumers aren't expected to generate as much economic activity. In the US, these effects are somewhat mitigated by having one of the largest Millennial cohorts so the labour market contraction isn't as drastic as elsewhere. In contrast, China is facing a demographic timebomb:

Based on whose statistics you’re using, the average Chinese citizen aged past the average American citizen sometime between 2017 and 2020. China’s labor force and overall population peaked in the 2010s. In the best-case scenario, the Chinese population in the year 2070 will be less than half of what it was in 2020. More recent data that’s leaked out of the Chinese census authority suggests that date may need to be pulled forward to 2050

"Geographies of success" and demographics are two themes that Peter Zeihan keeps coming back to. I think these factors are important too, so a lot of what he predicts from them should at least be taken into consideration. However, I don't buy that they are fully deterministic. Necessity is the mother of invention, so while things like energy might get a lot more expensive, I expect countries that are on the losing end of that change (i.e. energy importers) will find ways to tighten their belts and avoid collapse. On the other side, his case for the US coming out on top (no longer a global hegemon, but still enjoying peace and prosperity at home) doesn't wrestle enough with growing political division and animosity, in my opinion. Being self-sufficient in oil and gas is less beneficial if pipelines start getting shut down by state governors making a point, for example (let alone acts of sabotage). And things like the bussing around of illegal immigrants between states that we've seen in the past year seems like a harbinger of the sort of American Anarchy that Balaji describes.

One of the most valuable aspects of The End of the World is Just the Beginning is in arguing against the common view that China will be the next global hyperpower/hegemon. Zeihan may or may not be correct about this, but he identifies a lot of structural weaknesses in China (demographics, their financial system, their dependence as an export-driven economy (and importer of oil and food) on the rules-based international order they often seem to undermine) that anyone thinking about these topics should take into account.

It is also a very good strategic briefing when it is getting into specifics, like who has copper and where it is processed and used. And copper is only an example, he gets into the strategic geography of several key resources encompassing energy, minerals, and agriculture. For a lot of the metals, there is a recurring story where China is a massive importer, processer, and re-exporter:

The vast bulk of the world’s iron ore production comes from countries that face limited to no security threats as the world deglobalizes: in descending order, Australia, Brazil, India, South Africa, Canada, and the United States. However, the countries that export the vastness of global steel—in ascending order, Ukraine, Germany, Russia, Korea, Japan, and above all China—are somewhere on the sliding scale between “facing severe complications” and “utterly screwed.” The world is going to have massive shortages of steel, at the same time that the supplies of the raw material to make that steel will overflow.

As of 2021, China absorbs two-thirds of all internationally traded bauxite, while smelting about three-fifths of all aluminum. In true Chinese fashion, the majority of China’s aluminum output is almost immediately dumped on international markets. This is both great and awful. It’s great in that it simplifies understanding of the supply chains: China’s penchant for hyperfinancing and overbuilding makes it all China, all the time. It is awful in that the global supply chain for one of the world’s most utilized metals is wrapped up in a failing system. When China cracks, the world will face global shortages of aluminum, as there simply are not sufficient smelting facilities elsewhere to cover more than a few percentage points of the pending shortfall.

Along with the strategic geography of key resources, Zeihan offers reality checks on alternative energy and transportation.

Every technological development that has brought us to our industrialized, urbanized present must be reevaluated to make today’s greentech work. But by far the biggest challenge is the very existence of cities themselves. All are by definition densely populated, while greentech by definition is not dense. Squaring that circle even in sunny and windy locations will require massive infrastructure to bridge the gap between dense population patterns and far more dispersed greentech electricity-generating systems. Such infrastructure would be on a scale and of a scope that humanity has not yet attempted.

He is in favour of wind and solar power where the conditions are right, but wants to dispel illusions that the winding down of global trade in energy could be quickly or easily replaced with local sources. Similarly, international shipping has no substitute:

If you wanted a train that could compete in capacity with ships designed to just barely squeeze through the recently expanded Panama Canal, you’d need one more than forty miles long. Alternatively, you could go for a fleet of sixty-five hundred trucks.

As a vivid illustration of this, I looked up some statistics on China's Belt and Road Initiative. At the land port of Khorgas where the rail gauge changes between China and Eurasia, the annual cargo volume is 200,000 TEUs, which is only 0.5% of what the port of Shanghai (the largest in the world, to be fair) processes annually. (see this post for more about maritime trade).

For this reason, Zeihan's nightmare scenario for when the roll-back of globalization changes from a slow and steady trend to a rapid collapse is a commercial ship being sunk (as collateral damage of a conflict around the Hormuz Strait or the South China Sea, for example). According to him, the knock-on effects on insurance would immediately make international shipping a lot more expensive and curtail a lot of volume, with serious consequences to export-driven economies, just-in-time manufacturing, and globally-integrated production that relies on trade in intermediate goods.

everything we know about modern manufacturing ends the first time anyone shoots at a single commercial ship.

Contrasting Geopolitical Outlooks

One of the central geopolitical questions for the 2020s is on how the relative power and influence of China and the United States evolve. Of course, these are not the only countries—I'll cover other countries and regions in this section, don't worry—but understanding how someone thinks about this topic is a good starting point for their geopolitical outlook in general. Both Balaji Srinivasan and Peter Zeihan expect the US to pull back from being the global security guarantor, although they differ on the reason. For Balaji, it will be a consequence of declining state capacity (due to internal division), while for Zeihan it will be a political choice of isolationism as fewer and fewer Americans see the benefit of the current international order. As he wrote in an excerpt I included above, "the entire concept of the Order is that the United States disadvantages itself economically in order to purchase the loyalty of a global alliance," and with the end of the Cold War, he sees a growing, bipartisan decline in belief that this purchase is a good deal for the US anymore. However, it is in China that the starkest differences in their outlooks lie. Balaji expects the Chinese Communist Party (CCP) to tighten its control over the country. This won't be great for individual Chinese people, but the unity it brings will allow for the Chinese state to project power globally better than any competitors. In contrast, Zeihan thinks that China has some structural weaknesses that no one will be able to overcome and it will eventually splinter. Foremost among these weaknesses are demographics ("getting old before they get rich") and needing to import food, energy, and too many other critical raw materials.

(If I recall correctly, both authors include the famous opening line from Romance of the Three Kingdoms, though: 分久必合，合久必分 = "The empire, long divided, must unite; long united, must divide. Thus it has ever been.").

These views could be summarized as American Anarchy, CCP Control for Balaji (his words), and Fortress America(s), Chinese Fracture for Zeihan. I'll elaborate on what these views mean below, but first I want to note that laying it out like this allows for a wider possibility space to be conceptualized, as I mentioned in the Introduction:

American Anarchy, CCP Control
Fortress America(s), Chinese Fracture
American Anarchy, Chinese Fracture (things would get very chaotic both domestically and internationally under this scenario)
Fortress America(s), CCP Control (shoring up spheres of influence into a new Cold War)

Of course, to the above should be added a null hypothesis (scenario 0) that the status quo basically continues as it is. The complete possibility space also includes changes from current conditions in the direction that someone suggests, but of a lesser (or greater!) magnitude.

Regarding the USA, The Network State has this to say:

Both outside and inside the US, there’s the sense that the US-dominated postwar order is either on its last legs or already over, and that the ancient legislators and endless remakes reflect a fading culture trying to hang on by its fingernails to prevent what comes next.

Balaji also contrasts the US to China, writing:

The US establishment has the opposite set of constraints: unlike China, it doesn’t face organized military opposition abroad, so it’s highly incautious in its foreign policy. But also unlike the CCP it does face organized domestic political opposition at home, so it can’t be as ruthless domestically as it wants to be.

He sees the US edging towards some kind of Civil War, part 2. To be clear, this wouldn't necessarily be a shooting war, but perhaps a "civil cold war" or some kind of digital conflict between opposing subcultures and political factions. The reason for expecting this is partly from looking at current trends in political polarization and adding in potential economic difficulties:

All the political infighting of the last decade happened during a period of relative prosperity, even if it was based on the artificial expedient of printing money. But now that we face potentially years of inflation and stagnation, unhappiness will increase. Already you’re seeing articles coming trying to acclimatize people to lower standards of living, to “eat bugs and live in a pod.” And Turchin’s cliodynamical graphs put numbers to these feelings.

Cultural changes away from respect for institutions and hierarchies are also a factor:

The old American left said something like “we all need to work for the common good” while the old right said something like “pay your dues and you’ll achieve the American dream.” The new left says “we are all equal” and the new right says “you ain’t the boss of me.” So, the old left/right combination supported self-sacrifice and a stable hierarchy, while the new one attacks all hierarchy as fundamentally illegitimate, as oppressive or tyrannical. This is reflected in the defacement and degradation of virtually every US institution over the last few decades, from the office of the presidency to the statues of American founders.

Peter Zeihan also expects less US power projection abroad, but due to a democratic choice (on both sides of the aisle) for greater isolationism:

The post–Cold War era is possible only because of a lingering American commitment to a security paradigm that suspends geopolitical competition and subsidizes the global Order. With the Cold War security environment changed, it is a policy that no longer matches needs. What we all think of as normal is actually the most distorted moment in human history. That makes it incredibly fragile.

He also makes the point that the Americans went for a different approach to hegemony than traditional imperialism because (in the aftermath of World War 2), they had all the room they needed to expand on their own continent and because governing was neither their key interest nor core competence:

The Americans have never had a tradition of governing excellence because for much of their history they didn’t really need a government. Managing foreign territories twice the size of the United States would have been, like, really hard. And the Americans are, like, really bad at government.

I think this is a good counterpoint to Balaji's very negative outlook for the US. Good government has never really been the main thing that makes the US work and some level of anarchy isn't really a new thing for them—maybe not even necessarily a weakness. Though I'd say Zeihan is too sanguine that political dysfunction in the US won't negate some of the significant advantages it otherwise has (which he rightly identifies!—I agree that the US has a lot going for it and that being slightly ungovernable hasn't kept it down in the past).

More internal division in the US would favour Balaji's predictions over Zeihan's, though. Something like a state governor openly defying a federal law (like threatening to use the National Guard or state police to arrest federal law enforcement if they didn't back down) or like a ballot measure for secession appearing in one or more states (even if it didn't pass).

One big question is whether the US dollar will hold on as the global reserve currency. Zeihan doesn't see a good alternative, so even while the US retreats to a greater degree of isolationism, there won't be a new hegemon to take their place in a financial sense (or a military/security one either). A shift to other world currencies would line up more with the predictions in The Network State.

Regarding China, Balaji's emphasis is on their government's ability to accomplish its goals:

From 1978 to 2013, from Deng Xiaoping to Jiang Zemin to Hu Jintao, the CCP was focused on economic growth. But under Xi Jinping, it’s taken a turn towards militarist nationalism. It builds most of the world’s physical products, its military budget is already >1/3 that of America’s, it has a more focused task (“reunify China” rather than “police the world”), it produces military recruiting videos like We Will Always Be Here, and - most importantly - it is investing heavily in AI and drones.

The point about AI, also discussed above, is that it could offset some deficiencies in loyal manpower. His view that China will become stronger is in no way an endorsement of their methods, however. He thinks that the CCP will be increasingly authoritarian—and effective at this, thanks to their embrace of new technologies—not only over people in China, but over the entire Chinese diaspora:

A specific prediction is that we’ll see a world where it becomes increasingly difficult for Chinese people to leave the country or get their property out of the digital yuan ecosystem without CCP permission. Take the existing hukou system of internal passports, the WeChat system of red/yellow/green travel restrictions based on health status, the aggressive COVID lockdowns, and the recent passport restrictions — then fuse them with a surveillance state that can track people globally, a WeChat superapp that can unperson them, and a digital yuan that can freeze their assets.

Zeihan is far less impressed by the Chinese government's managerial abilities. He claims that even things that look like strengths, like their high industrial production, are a symptom of their economic system not being as good as a market economy at matching supply and demand:

Another result is massive overproduction. China is worried about idle hands, not bottom lines. China is by far the world’s largest exporter of steel and aluminum and cement because it produces more of all three than even hyper-ravenous China can use. China’s much-discussed One Belt One Road global infrastructure program—which many non-Chinese fear is part influence peddling, part strategic gambit—is in many ways little more than a means of disposing of the surpluses.

Obviously Zeihan puts a lot more weight on fundamentals like demographics and secure access to critical resources (importing oil from 10,000 km away through a couple of vulnerable chokepoints is anything but secure!), and thinks China is headed for turmoil on this basis. Whereas Balaji's view is that emerging technologies will give the CCP the state capacity to maintain control. Indeed, the second half of 2022 has seen some crises start to pile up in China, ranging from financial (bank runs and real estate bubbles, not that they are unique in the latter), to ones related to natural resources (drought leading to black-outs as hydropower dries up, very much related to a recent post here), to rare public displays of discontent. It's at least possible that these crises (especially if they persist instead of turning out to be rather transient) are related to some of the structural weaknesses Zeihan sees in China. However, so far Chairman Xi still seems to be firmly in control, so I don't think we can judge which author is closer to the mark yet.

Just like Balaji underrates the importance of population (and agriculture, I'd say), I think Zeihan has too static a view of force projection. Right now, he lists the US, Russia, UK, France, Turkey, Japan as the only nations that can operate navies far enough afield to secure resources. But how can he be so sure other countries (obviously China, but also others like Korea which has tons of shipbuilding) won't develop those capabilities if the need is great enough?

Regarding Russia, I don't recall that Balaji's book had much in the way of specific comments or predictions. Zeihan's book clearly had some edits that were made just before publication to touch on the invasion of Ukraine. In spite of having an almost-terminal demographic outlook, he thinks Russia has enough other strengths that they can't be counted out:

Russia is probably one of the best-case scenarios for much of the industrialized world. Russia, after all, at least has ample capacity at home to feed and fuel itself in addition to sufficient nuclear weapons to make any would-be aggressor stop and think (a few dozen times) before launching an assault. In a world of constrained trade and capital, one could be in significantly more dire straits than still having strategic depth plus reasonably reliable food, fuel, and electricity.

On this point, note how many Russian oligarchs derive their fortunes from strategic essentials like iron/steel, oil & gas, and fertilizer.

Bruno Maçães' Dawn of Eurasia offers yet another view of the prospects for Russia and the surrounding region. If there is a continued shift of the global economic center of gravity to Asia, while Europe remains a strong consumer market, then sitting astride a land-based trade route between them and having plentiful energy and mineral resources to sell both east and west will be a very strong strategic position. Build-out of the Belt and Road infrastructure would also mitigate at least some of the weaknesses Zeihan sees in China, as they could access a lot of what they need overland through territory they control. However, if I understand correctly, Zeihan would disagree with Maçães on a few points: he doesn't think east Asian economic growth is sustainable (too indebted/leveraged), he doesn't think European demographics will maintain consumer demand, and the efficiency drop from water transport to road and rail means that the latter isn't a substitute for the former if geopolitical chaos imperils ocean-going trade (as noted above in the example of Khorgas).

Other countries I won't spend as long on as the superpowers in this review. However, I think it's worth sharing some of these authors predictions for various countries and regions to be able to evaluate the accuracy of their models as events unfold. Because Peter Zeihan expects a trend of growing isolationism from the American political system and less global security as a result, he predicts a decline in global trade. Decades of globalization has allowed for a lot of specialization and gains from trade, but also a lot of dependence on its continuation; some places are more exposed than others, though:

It begs the question of what Shanghai would look like without oil? Or Berlin without steel? Riyadh without . . . food? Deglobalization doesn’t simply mean a darker, poorer world, it means something far worse. An unraveling.

Places that are likely to do well (at least relatively—most will be at least somewhat poorer if these things come to pass), are those that are close to self-sufficient in vital resources and self-defense. First of all, Fortress North America (with Colombian/Chilean/Argentinian and maybe Australian annexes) is pretty viable in Zeihan's analysis:

Within North America as a unit, more than 8 in 10 dollars (or pesos) of income is generated within the continent. That’s by far the most insulated system in the world.

In Europe, the old quip about NATO was that its purpose was to "keep the Russians out, the Americans in, and the Germans down." Because he expects a US retrenchment, he also expects a less united Europe:

The United Kingdom, France, Germany, Sweden, and Turkey will all go their own way and attempt to attract and/or coerce select neighbors to come along for the ride.

One of Balaji's themes is that the nation-state model is becoming obsolete, so he doesn't get as much into analyzing the prospects of existing countries. His book provides a useful reminder that a lot of non-state actors (e.g. corporations, cryptocurrencies, organized social networks) should be included when thinking about geopolitics. He also considers that non-aligned countries will have a vital niche to fill between his predictions for American Anarchy and CCP Control. India is one of the foremost examples of a country that is well-positioned to chart its own course. Their improving infrastructure is one example of the strength of their position. Even more than India the country, Balaji says he's bullish on the Indian diaspora.

Before wrapping up this section, I want to discuss manufacturing and agriculture. The stochastic supply chain challenges of the past few years need to be taken into account in any forward-looking planning and predictions. Zeihan certainly sees this as the new normal:

The globalization game is not simply ending. It is already over. Most countries will never return to the degree of stability or growth they experienced in 2019. And now most have lost the chance to even try to shift onto a newer, more appropriate footing.

The sheer fluidity of the future security environment won’t help. The industrial plant required to support multi-step supply chains exists in multiple locations by definition, and takes years to build. Every time there’s a tweak to a demand profile—either for intermediate or finished goods—it typically takes a year of retooling efforts to work its way forward and back through the system. We have learned that little lesson the hard way with COVID. Every ship diverted, every shot fired disrupts some part of the supply and forces that same year-long reset. In such an environment, multi-step supply chains in any region without rock-solid local security and rock-solid local consumption just don’t make much sense.

This makes things like machine shops and 3D printing more advantageous:

The new systems will put premiums on simplicity and security just as the old system put premiums on cost and efficiency. The death of just-in-time will force manufacturers to do one of two things. Option A is to warehouse masses of product—including finished product—as far forward in the manufacturing process as possible, preferably at the very edge of major population centers. Option B is to abandon as much of the traditional manufacturing process as possible and do all-in manufacturing as physically close to the end consumer as possible. One technology suited to the latter is additive or 3-D manufacturing,

3D printing also fits well with the worldview of The Network State. It envisions a future where the continued advance of technology has a lot more things digital first even if they ultimately get materialized in the physical world. Balaji expects more manufacturing to be automated and on-demand. This is also well suited for network states that don't have a lot of physical territory since it can have a very efficient physical footprint.

On the other hand, I don't think Balaji has enough appreciation for the importance of agriculture. A network state with an archipelago of small, scattered physical territory (he specifically says some of these territories could be merely a cul-de-sac or even smaller) might work for digital nomads who can work from anywhere, but it's not going to be an exporter of agricultural commodities. And advances in biotech and efforts towards a circular economy are poised to make agriculture even more important, not just for food but for materials and perhaps energy (for one example, the market for polylactic acid, a bio-plastic, is growing by 10-15% per year). Zeihan, by contrast, sees agriculture as being of existential importance:

The first category of food-exporting countries are those whose supply systems for everything from finance to fertilizers to fuels are sufficiently in-house that they can continue producing their current product set with only minor adjustments. France, the United States, and Canada are the only countries on the planet that check all the boxes. Russia is a near miss.

At the end of the day, it was valuable to read these two books back-to-back. Reading one of them in isolation probably would have been more convincing, but hearing both author's arguments gave me space to weigh the strengths and weaknesses of their assumptions and also contemplate other possible scenarios as I've tried to explain above. God only knows what the future actually holds, so rather than treating either book as an authoritative geopolitical forecast, I'm treating them as potential scenarios to contemplate and permutate in my own planning (for things like investments, career opportunities that require relocation, etc.). Combining their scenarios, it shouldn't be unthinkable (nor do I think it's likely, just not outside the realm of possibility) for any major power of the moment to unravel to some degree: China, Russia, the US (e.g. certain states going their own way to the point that there are internal border controls), or the EU (e.g. countries ditching the Euro or leaving the Schengen free movement area—Brexit was about the UK leaving the EU, but they weren't part of the Euro or Schengen to begin with, so it isn't necessarily the start of a trend the way a truly core member leaving would be). At the moment, there are serious signs of division in the US (e.g. taking the longest to elect a Speaker of the House since the pre-Civil War days), China is dealing with a serious Covid wave after their zero-Covid approach became no longer tenable, Russia is looking weak in Ukraine, and the EU has been facing an energy crunch (made slightly easier by a mild winter). So nowhere is really looking like it will win at the moment. On the other hand, I don't think it's wise to underestimate the human propensity to muddle through in times of necessity, or the sheer inertia of the status quo.

Miscellaneous Material

This section is for material from The Network State and The End of the World is Just the Beginning that didn't really fit anywhere else in this post, but was too fun or interesting to leave out.

One of Balaji's examples of a moral innovation ("One Commandment") to build community around is the idea of a digital Sabbath (previously discussed here):

One way of accomplishing this would be a Digital Sabbath society where the internet is just shut off at night, from 9pm to 9am. Some buildings and rooms would furthermore be enclosed in Faraday cages, to put them offline on purpose. Areas would start to be flagged as online and offline areas, a bit like smoking and non-smoking areas on planes. All internet use would be conscious and focused, as opposed to unconscious and involuntary.

I actually saw in the news recently that a village in India has sort of tried this (for 1.5 hours per day).

Zeihan's book had a bit to say about the Mekong River, the topic of one of my posts from last year:

Among the world’s major rivers, the one that has seen the greatest changes to its volume and flows in recent years is the Mekong, in Southeast Asia. The Chinese have tapped its upper reaches to irrigate fields on the Tibetan Plateau, the Laotians and Thais are building dams like mad to generate hydropower, the Cambodians have centered their civilization on the intersection of the Mekong and their seasonally flooded lowlands, while the Vietnamese have turned the Mekong’s entire delta into one gigantic rice paddy. Deltas being where rivers meet ocean, you can see the problem. Even marginally lower river flows lead to both the land sinking a bit and the sea pushing inland.

I also found this paragraph from his book very amusing:

If you ever find yourself needing to stress test an energy-related theory, Vaclav Smil of the University of Manitoba serves as a one-stop shop. That’s not quite right. The guy has written more books on the reality of energy than I have socks, and my sock game handily exceeds that of the Canadian prime minister.

In fact, I'd say there are some major similarities with Smil's style in terms of analyses rooted in fundamentals of available energy and material resources, (although Zeihan's writing style is much more conversational and even flippant in places). An example of this type of analysis is some of his points about renewable energy and the large footprint it requires. This increases the hinterland area that a city needs to tap into, or, conversely, decreases the sustainable population density. This makes me think of a point I heard on a podcast (I believe it was an interview with Alain Bertaud) point about the biggest value of cities being the way they bring a large and diverse labour market together: workers have more bargaining power and a chance to specialize and switch to a role they're best suited for; employers have a good choice of qualified candidates. The key is the size of the catchment area (determined by transportation technology for a reasonable commute) times its density. Zeihan makes the point that the larger the city, the more dependent it is on stable, global trade: to draw resources (including food) from further afield, and to support that high a degree of specialization:

Modern cities are so large and their economies so specialized that they require regular access not just to a huge swath of territory, but to the entire globe.

But Balaji's focus on technological change would probably counter that the rise of remote work might make it easier for people to physically spread out while still getting some of those advantages of a large integrated labour market. A synthesis of these views might be a future that is more digitally interconnected, but physically disconnected (e.g. intercontinental VR meetings and even virtual "tourism"); it's not impossible that we could get a degree of deglobalization in physical goods (at least so far as routine bulk trade goes) while media and digital goods & services can still go international (perhaps with some personal travel and small packages that can be delivered by plane continuing as well). That is,9 deglobalization colliding with dematerialization trends.