Mancur Olson’s The Rise and Fall of Nations does not offer a particularly optimistic picture. Once a nation has stabilized for a while and even grown wealthy, it becomes increasingly vulnerable to “institutional sclerosis.” This happens because small groups can successfully overcome free riding and thus effectively bias the system in their favor. The whole system deteriorates as more and more of these groups emerge and are able to survive and benefit from the competition that brings about general progress and growth.
Taking Olson’s research to its logical conclusion, a highly effective cure for economic stagnation is devastating war. It’s clearly not the desired solution. But Olson was pointing out a real problem. The longer a society remains stable, the more likely it is to be suffocated by special interest groups. These “distribution coalitions” are not interested in expanding the economic pie. They just want to use government to protect and grow their wealth. Over time, their relentless rent-seeking hardened the entire system, Olson noted, and historically it took a major shock, like the complete devastation of Germany and Japan in World War II, to wipe out the “white slate of the system.” Stripped of trained lobbyists, these countries were in a terrible position to unleash economic growth.
But relying on system collapse or war to wipe out rent-seekers is clearly not a viable policy prescription. Achieving this Olsonian “institutional clean slate” requires peaceful mechanisms, and this is where artificial intelligence comes in as a potential system shock.
To understand how this mechanism works, let’s apply some systems thinking to a concrete example. It’s the notoriously complex German tax system. Currently, the high density of Germany’s tax law acts as an artificial barrier to entry, creating huge rents for a particular distributive coalition: tax consultants, bureaucrats responsible for overseeing taxes, and politicians who can distribute rents to advantaged groups. These groups are in an advantageous position because navigating the bureaucratic maze requires highly specialized human resources. As a result, they have strong incentives to lobby against meaningful tax simplification. Because doing so would destroy their business model.
Artificial intelligence provides an exogenous technological shock that could shatter this stasis. If AI can parse and execute complex tax codes at a fraction of the cost, the economic foundation of the tax consulting industry will effectively disappear. As the sector’s revenues dry up, its financial capacity to fund lobbying efforts simultaneously shrinks. Without a powerful and well-funded rent-seeking group that actively demands that the tax system remain complex, the political frictions that impede reform will persist. In this scenario, the technology would come back to a clean slate, the coalition’s lobbying power would be greatly reduced, and meaningful legal reform would ultimately be possible.
But not only does this sound too easy to be true, it overlooks one thing: the resilience of strong coalitions. Olson specifically pointed out that distributive coalitions inherently seek to slow society’s ability to adopt new technologies in order to protect the status quo. Existing industries should be expected to engage in creative rent-seeking before AI completely erodes their lobbying power. For example, tax coalitions are very likely to lobby governments to require that AI-generated tax returns remain legally invalid unless reviewed and stamped by accredited human experts, citing common justifications such as “data privacy” and “liability.” In other words, those who profit from rent-seeking may, for example, hide their own interests with the noble defense of “tax justice” or raise the risk of “algorithmic bias.” Rather than quietly withering away, the existing coalition is likely to devote all its resources to a final, intense lobbying effort to regulate the presence of AI before it expands. This inevitable backlash will lead to difficult political battles in the coming years.
Winning this battle requires recognizing that a broader political economy is at play. If we want to escape the trap of institutional sclerosis, we need to understand that AI is not a silver bullet. Rather, it is essentially a window of opportunity to weaken and overcome strong distributive coalitions. But to take advantage of this window to unleash Schumpeterian destruction, we must fiercely resist these coalitions that seek to regulate and eliminate these new technologies from existence. This could include, for example, raising awareness among the population as a whole, something that is notoriously difficult to organize in large groups. This will certainly be a challenge for economists and Hayekian ideas secondhand dealers who consciously protect innovation and freedom to innovate against rent-seekers who want to protect their rents. What is certain in all of this is that the battle for a clean slate is not automatically won. Action is needed on our part.
