The Ratchet Effect

Regulation is the implementation of distrust.

Imagine a world in which everyone could trust everyone else with perfect certainty to know the right thing to do and to want to do it. This is the libertarian fantasy that Daniel Kahneman summarized in Thinking, Fast and Slow.

“The assumption that agents are rational provides the intellectual foundation for the libertarian approach to public policy: do not interfere with the individual’s right to choose, unless the choices harm others.”

The rational person knows all and understands the consequences of their conduct. They will not do things that harm others. The rationality and the social sensitivity of the individual is what sustains their freedom from interference.

It is sufficient to intervene if either of these conditions fails.

An economist would say that man responds to incentives. Perhaps we can think of life as a multiplayer, infinite-lived game. Over time, it pays to collaborate, so that everyone is better off even as there will be benefits for the individual from cheating by acting selfishly to obtain a greater payout in any given period.

Regulation is therefore a way to skew the rewards, penalizing offensive individualistic behavior. It changes the incentives to make things better for everyone, ideally without hurting anyone in the process.

For example, consider ocean fishing. In an unregulated context, it makes sense for the individual to maximize their welfare by catching as many fish as they can. Extrapolating that logic to a large number of people suggests the likely outcome to be overexploitation. In the short term, the individual is better off. In the medium to long term, everyone suffers from the permanent depletion of the stock, while they wait for the population to recover.

There is a dynamic aspect to our multi-player, infinite-lived game. We’d like to smooth the set of outcomes across time, to maximize the aggregate of individual utilities.

Governments or groups issue rules. Individuals bear the risk and the cost (including proving compliance) so that the collective can benefit. There is a disconnect between who bears the pain and who receives the offsetting upside. The regulator does not pay any price.

The incentive for the rule maker is always, and in every case, to constrain individual actors, be they people or groups. They have no apparent disadvantage. It is like a ratchet. Prescriptions on behavior and the attendant sanctions for non-conformity tend to move in one direction: up. If you like one rule, you’ve got to like five.

From a social perspective, abstracting from the question of who bears the costs by considering only the social impact, one can imagine an optimal level of intervention. When there is too little, the marginal social benefit of an additional rule exceeds its marginal social cost. Conversely, with too much, the opposite holds true

There is a best case, one that is neither too hot nor too cold.

But government claws back regulation only when it is deemed to have gone too far, a conclusion driven as much by culture and history as by rational analysis. It is daunting to rollback regulation because of the bureaucratic ossification that results from rendering these prescriptions. People have jobs enforcing and administering the controls, monitoring compliance, and imposing penalties. Institutions become dependent on fees. The population alters their behavior and expectations to conform to the artificial conditions that overburden a specific type of activity.

Every new rule is a marginal transfer of control from the individual agent to the bureaucracy. Where there is no regulation, there is perfect freedom. There is no such thing in a modern society. There are only degrees of autonomy. Control here is the power to direct the use of a resource or to mandate behavior.

The social contract is a fabric of myriad such concessions by individual agents in which they relinquish some aspect of their personal power in exchange for a peaceful social order. Each of these regulations is the micropolitical tradeoff that rolls up into a set of social goods including, ideally, peace, order, and opportunity for self-actualization. No compromise is perfect and this tapestry of interaction evolves continuously in uncountable dimensions.

The regulator’s imperative to add to its power in a manner that comes at the expense of other players is almost impossible to counter organically. It requires either an authority able and willing to identify principally with the interests of individual actors or a set of conditions that present no alternative but liberalization.

The cumulative ratchet effect has to become so great that there is a political benefit from implementing deregulation or there is a tear in the social fabric which makes the willing tradeoff of control for order untenable. In either case, the bureaucracy must have reached a level at which it has poisoned the well. People look for a cleansing reset, if only a partial one.

One recent example is the impact of the Chernobyl disaster in accelerating the decline of Communist rule in the Soviet Union.

“Both policies [glasnost and perestroika] link to the concept of a risk society, as the Soviet Union were exposed of their selfish agendas behind Chernobyl. The selfish priority of the economic well-being of the Union and the silence on the matters of risk to the inhabitants creates an environment that provides a perfect example of a risk society, where the distribution of wealth is of greater importance than the distribution of risk. Therefore, Chernobyl did challenge the fulfillment of the policies of Gorbachev, which prevented the necessary reformation of the governmental system and contributed heavily to the downfall of the Soviet Union.”

Another connected example, from the same period in recent history, is the Indian financial crisis of 1991. Faced with a vicious spiral of economic events including spiking oil prices due to the Gulf War, credit ratings downgrades, the inability to pass a budget, and the eventual suspension of IMF and World Bank assistance to the country, India was forced to accept terms from these global financial institutions.

Crucially, these required reform of India’s infamous “License Raj.”

“Up to 80 government agencies had to be satisfied before private companies could produce something and, if granted, the government would regulate production.”

Before reforms, the government had almost complete control; companies existed at the whim of the bureaucracy. Naturally, this was an invitation to concentration and corruption.

“One consequence of the License Raj was that it benefited large corporations at the expense of smaller businesses. Because large corporations were often better able to navigate the complex bureaucracy of the License Raj and secure the necessary licenses, they were able to dominate many sectors of the economy. This made it difficult for small businesses to compete, and contributed to a concentration of economic power in the hands of a few large corporations.

“Another criticism of the licensing system in India was that it was prone to corruption, as businesses and individuals had to navigate a complex bureaucracy in order to obtain licenses and permissions, and may have had to pay bribes or engage in other forms of corruption in order to obtain the necessary approvals. The corruption was fueled by a broader environment f corruption in India, which was characterized by a lack of transparency and accountability in the government, a weak legal system, and a culture of corruption that had been allowed to persist for many years.”

There is an apocryphal story to explain the explosive growth in the IT sector in India. As it is told, the bureaucrat responsible for overseeing this industry was so short-sighted and distracted he dismissed its potential and could not be bothered to regulate it. As it became so large, so quickly, it became politically impossible to put the cat back in the bag. The economic value was too high. True or not, the story is evocative.

The Indian IT sector is now 7.4% of GDP.

No wonder the country is hesitant to regulate artificial intelligence.

“In a statement on Tuesday, India’s Ministry of Electronics and Information Technology acknowledged numerous ethical concerns around bias and transparency that could arise with AI’s rapid expansion but explicitly said the Indian government ‘is not considering bringing a law or regulating the growth of artificial intelligence in the country.’ The ministry instead referred to AI as a ‘kinetic enabler of the digital economy,’ which it believes will strengthen entrepreneurship and business and play an important strategic role for the country moving forward.”

This stands in contrast to the hysterical reaction in the West to the apparently sudden emergence of generative AI. Large companies and government officials proselytized recently for extensive regulation (which, coincidentally, would protect the positions of incumbents).

“When it came to how government should regulate AI – and LLMs [Large Language Models] in particular – Altman [Sam Altman, OpenAI] suggested that the government should focus on safety requirements before and after testing, working with other governments to agree on a common approach to licensing and auditing these AI models.”

The License Raj emerges in the West.

Advocates argue that the justification for AI regulation is uncertainty. We cannot trust individuals to know how to act properly. The economic incentives are too great to rely upon individual actors to act in a manner that does not damage the social fabric, already fraught from the strain of polarization. We cannot know what will happen, so we must assume the worst.

The ratchet theory of regulation stipulates that regulation rises monotonically to a point of political and/or practical failure, driven by insensitivity to the incidence and distribution of risks and compliance costs such rulemaking imposes and by the bureaucratic imperative to accumulate control for the power it conveys.

What then is the natural antidote, if any? How can individual actors evince the trust and expertise necessary to forestalling restrictions of their behavior? Short of demonstrating the economic exigencies of competing on the world stage with countries that are willing to invest in deregulation for the economic growth it begets, there may not be much. The ratchet effect predicts cycles of regulation, failure, and liberalization.