The argument runs like this. Modern economies are not really run by governments or markets but by an energy-hungry system that behaves like a single organism, one that grew on a one-time inheritance of fossil fuel and cannot keep growing once that inheritance thins out. What follows is not a crash but a simplification: less energy per person, shorter and more local supply chains, smaller and less complicated ways of living.

That is the core of The Great Simplification, the framework popularized by Nate Hagens through his podcast and writing of the same name, built around ideas he calls the carbon pulse, energy blindness and the superorganism.

It is worth being clear about what kind of claim this is. It is a synthesis presented through a podcast and an education platform rather than a peer-reviewed research program, though the superorganism concept itself has a peer-reviewed paper behind it, Hagens writing in Ecological Economics in 2020. And it makes a forecast, which no dataset can confirm in advance. But parts of it rest on published work that can be checked, and parts of it make predictions about the present that can be tested against current data. Those two things point in different directions, which is the interesting result.

The part with peer-reviewed support

The intellectual ancestry here is real and older than the podcast. Joseph Tainter’s The Collapse of Complex Societies, published by Cambridge University Press in 1988, argued that societies grow more complex in order to solve problems, that complexity has to be paid for in energy, and that investment in complexity eventually reaches a point of declining marginal returns.

The modeling strand traces to the 1972 Limits to Growth study, and it has been revisited recently in a peer-reviewed venue. Gaya Herrington’s update comparing the World3 model with empirical data, in the Journal of Industrial Ecology in 2021, tested how far observed data tracked four of the original scenarios.

Her finding is more specific than it is usually reported. The two scenarios that aligned most closely with the data both indicate a halt in welfare, food and industrial production over the next decade or so, and both indicate subsequent declines. But only one of the two, the one where declines are caused by pollution, depicts a collapse. A halt in growth and a collapse are different outcomes, and the data as of that analysis did not distinguish between them.

A separate group revisited the model again. Arjuna Nebel and colleagues published a recalibration of World3 in the same journal in 2024, tuning the input parameters to better match world development data. Their recalibrated run shows the same overshoot and collapse mode as the original business-as-usual scenario, with the main effect of the update being to raise the peaks of most variables and push them a few years further out.

That is a genuine result and it deserves to be reported as one. An independent recalibration reproducing the qualitative behavior of a fifty-year-old model is not nothing. It is also a model, tuned to fit past data, and a model reproducing its own structure under new parameters is weaker evidence than it can appear.

What the trade data actually show

The localization half of the thesis makes a claim about the present, and that can be checked directly.

The usual number offered in support is that world trade peaked as a share of global output in 2008 and has been receding since. In the current World Bank series it did not.

Trade as a share of world GDP was 25.8 percent in 1970 and 60.3 percent in 2008. It then dipped, recovered, and reached 62.1 percent in 2022, higher than the 2008 figure that gets quoted as the high-water mark. It fell to 58.1 percent in 2023 and 56.7 percent in 2024.

So there is a real decline in the most recent data, and it is two years long. Two years is an observation, not a trend, and the same series fell in 2019 and 2020 before rebounding to a record. The World Bank has also posted a 2025 figure of 68.5 percent, a jump of nearly 12 points in a single year. A move that size in world trade openness is not plausible as a real annual change, and it most likely reflects incomplete country coverage in a newly posted estimate, though the database does not say so. We are not treating it as usable.

Reallocation is not localization

The stronger test is not how much trade there is but where production is going. If the simplification thesis is right, activity should be moving closer to home.

Cody Kallen of the Federal Reserve Board examined this in an April 2025 note on geopolitical fragmentation and US foreign direct investment. Outward investment has shifted away from China and Hong Kong and toward Mexico, India and Vietnam, though the Vietnamese share remains small, and US multinationals have moved expenditure and employment along the same path. Mexico, the largest gainer, is nearer to the United States than the destinations being left behind, so this is nearshoring in the literal sense.

On the question of whether any of it is coming home, the note is blunt. It reports little evidence of higher domestic investment and production shares by US multinationals overall as of 2022, and describes only tentative signs of reshoring in high-tech industries and advanced manufacturing, not more broadly. It also cautions that a rise in the domestic share of capital spending among high-tech firms could reflect data center and AI investment rather than reshoring.

The International Monetary Fund reached a compatible conclusion in its April 2023 World Economic Outlook. Chapter 4, on geoeconomic fragmentation and foreign direct investment, found that post-pandemic foreign direct investment declined by almost 20 percent against the pre-pandemic average, but that the decline was extremely uneven, producing relative winners and losers rather than a general retreat homeward. Its modeling put the long-term cost of investment fragmentation at around 2 percent of global output.

Two percent of global output is a serious economic cost. It is not a simplification of the kind the thesis describes.

Where that leaves the argument

Separating the two halves is the useful move here.

The energy and complexity half has a real lineage, a peer-reviewed modeling literature, and an independent recalibration that reproduces the original behavior. It concerns decades, and no current dataset can settle it either way.

The localization half makes a claim about now, and the current evidence does not support it. Trade openness did not peak in 2008, it reached a higher level in 2022, and the recent fall is two years old. US production, on the best available evidence, is not returning home. It is moving between foreign countries on political rather than economic lines, which raises costs without reducing overall foreign dependence.

Fragmentation and simplification are being treated as the same phenomenon, and on the available evidence they are not. The world is rearranging which distant countries it depends on. That is a different thing from needing fewer of them.