Towards a liberal autocracy

Why classical liberalism is incompatible with democracy

China is democratic. Yes, you read that right. China is a democratic state. This is not just a statement of CCP propaganda, but reflects a sophisticated history and philosophy of how Chinese institutions have operated throughout much of its civilisation.

Officially, China is a “people's democracy”, or 人民民主 (rénmín mínzhŭ)¹. The 共产党 (Gòngchăndăng) has a presence across all of civic life and any organisation sufficiently large. Their mass membership model ensures the party is relatively representative, albeit skewed towards the right-tail of IQ and conscientiousness given the highly selective process involved to become a member. This allows the party to legislate via a consultative approach, and indeed arguably the CCP holds more information on the desires of the median and modal citizen than most electoral representatives in the West. As a result, the CCP is highly responsive to public opinion, which explains why they imposed zero-Covid lockdowns with delayed vaccine rollouts (and their refusal to use the more effective Western vaccines) for as long as they did.

In fact, it seems to me that the de-facto ruling ideology of the CCP is Confucianism dressed up in socialist rhetoric. At its core, the CCP promotes a 小康社会 (Xiăokāng Shèhuì), a “moderately prosperous society” that seeks to balance economic growth and social stability². Harmony is a central virtue of Confucian philosophy.

Under Confucianism, the state has traditionally derived a “mandate from heaven”, or 天命 (tiānmìng), as its basis of legitimacy. The leader obtains the moral right to govern via promoting moral virtue in his actions and throughout society. If the leader however falls short of these standards, or fails to maintain social stability (a moral value in Confucianism in itself), then the people reserve a “right to rebellion” to violently overthrow the ruling regime. Indeed this explains the political turmoil of much of Chinese history, such as the overthrow of dynasties and the civil war. In fact, after the disasters of the Great Leap Forward and the Cultural Revolution, the party faced a mounting challenge to its legitimacy. To maintain power, the ruling elites essentially had to enact a full-reversal yet without being explicit about it and self-undermining its legitimacy (explaining why socialist rhetoric remains to this day). Even this wasn't sufficient per se, with serious and credible proposals to reform towards a more Western-style liberal democracy that only ended with the failure of the Tiananmen Square protests.

Most Westerners tend to (at best) yield the most cursory and superficial understanding of Confucianism, which leads to much misunderstanding across both sides. A Confucian understanding of democracy is rather distinct. Broadly, one can define democracy as governance “by the people”, yet this needn't involve elections. Rousseau, arguably the inspiration for Western populist movements today, long emphasised this. Similarly, Confucianism in effect invokes the Roussean approach, and this is crucial for the survival of the CCP today. Hence, if a state as authoritarian as China can credibly claim to be democratic, what does this imply for the normative implications of democracy? Should we reassess our views on its relative merits?

I argued in that piece that the romantic approach of submission to the common will imposes a contradiction to democracy via undermining political pluralism. Yet I made a basic category error which most writers make. Democracy is not identical to liberalism. Democracy is fully consistent with authoritarianism, and indeed given the preferences of the median voter across the planet, democracy implies authoritarianism in practice. Liberalism advocates a democracy with checks and balances to sometimes constrain the majority will. However most “liberal democracies” impose policies that sharply depart from liberalism: both the philosophy and what most self-described liberals advocate.

Increasingly, it’s obvious to me that (classical) liberalism is mutually incompatible with democracy. My definition emphasises a permisssive stance on all behaviours that don't impose direct and proven harms on others, under the governance of a state that protects property rights and curtails aggression. As I write this, my government, in a relatively free country, is proposing price caps for major grocery chains. Rent control, not only an illiberal policy but the most economically illiterate policy out there, is being supported by highly influential figures on the left, and could well be official policy. NIMBYism is a prime example of excessive democracy constraining private markets and free interaction, and is a major source of our economic stagnation.

Immigration restrictions are a particular pet peeve of mine: I consider it a strong moral obligation to adopt open borders. Allowing people to escape the poorest and most war-torn places on the planet to the most prosperous would dwarf any philanthropic intervention in terms of the gains to global wellbeing. Open borders would undoubtedly be one of the most effective and consequential means to reduce suffering today. Extreme poverty in the third-world is largely preventable via this mechanism, yet given our refusal to open ourselves, we choose to prolong such unnecessary misery today because our electorates want closed borders. The most lax policies towards immigration are overwhelmingly adopted by autocracies such as the UAE. This means that from a utilitarian perspective, democratic elections preventing the free movement of individuals to the richest places result in democracy being bad for welfare.

Previously I came round to a lukewarm support for “liberal democracy” on the Hayekian argument that decentralised institutions are superior for information aggregation. Elections encode the preferences of the public in a means that no centralised social planner can, and this facilitates accountability. Augmenting with a Burkean outlook, the calculation problem implies that spontaneous social norms arising organically via “trial and error” are optimal for inducing cooperation, and encode the wisdom of the centuries. However, Arrow proved explicitly that no aggregation rule can be guaranteed to reflect the preferences of citizens on policy outcomes, therefore elections are useless from an accountability perspective.

I now believe that liberalism and the Enlightenment can only survive via an autocracy, although this is what many of the original liberal thinkers also believed. Indeed the French Revolution drew some opposition amongst liberals for this very reason. Rather than absolute monarchy, ny ideal solution is a libertarian one-party state with a constitution and internal processes or checks to constrain the worst instincts of authority, yet a state constrained by basic classical liberal principles and holding a political monopoly on power. Nonetheless, those early liberals were correct. Why would one necessarily think the masses support basic freedoms when the evidence suggests they clearly don’t? For liberal defenders of democracy, how can you possibly tolerate a tyranny of the majority via elections imposing restrictions on the liberties of others?

A core tenet of liberalism is that government must maintain legitimacy - the consent of the people to govern. On this, all the discussed approaches agree, with the disagreement being on whether this is best achieved via pandering to popular opinion. Legitimacy can be achieved via ensuring the efficient allocation is implemented, which generally implies free-market capitalism. People can vote with their feet into and out of the country. Freedom of speech and expression can be maintained, as one needn't criminalise expressing discontent with this liberal autocracy; only concerted and coordinated action to overthrow the regime would be crimimalised. This is basically our approach regarding terrorism, which is ideological at heart. Overall, one would be freer under this proposal than in any democracy currently existing. If all nation states adopted this, genuine political competition and pluralism will arise across countries (rather than within) anyhow. With a clear mechanism to obtain legitimacy, one can preserve the benefits of democracy in reducing the incentives for civil war. This is how almost all companies operate, so why would we expect that governments are more effectively managed via elections as opposed to a CEO?

Empirically, there appears to be little observable or causal difference in the economic performance of autocracies vs democracies (using the standard classification approaches). Cross-country comparisons show a high correlation between economic freedom (and GDP) vs democracy, yet establishing causality is harder. Even if such a natural experiment arose, one would almost certainly require synthetic control to stimulate a likely counterfactual (parallel trends simply don’t exist for countries), which is arguably inherently non-robust as such counterfactuals often depend on the specifications of the author. However, my priors are that democracies minimise variance, which is why most elites support them despite tending to depart from elite preferences in implementing policy outcomes. Instead, I consider the populist vs elite (as defined via elite human capital theory) distinction³ to be far more consequential in determining the long-run economic performance of a nation, regardless of whether such is democratic or autocratic. Democracies tend to cycle between populist vs elite or technocratic regimes however which perhaps provides a stabilising effect.

Regardless, why should we be constrained via what has been observed up to today? The early Renaissance intellectuals dared to dream, and improve institutions to the far-right tail of performance at the time, and we're reaping the dividends of that bold bet to this day. Just because many autocracies are in the worst tail of governance does not imply that we can't improve on policy outcomes today by innovating from democracy.

1. I’m learning Mandarin so writing in logographic form (I've always referred to this as Kanji yet I recently learnt that’s the Japanese term for Chinese script - I studied 日本語 in secondary school). This helps me memorise the characters. Yes these are technical terms but many of the characters seem to occur in common phrases. It seems that each character corresponds to one syllable, each often a word in 中文, and by aggregating these together you get the English equivalent of a word.

2. This is why 习近平, Xi Jinping, has recently deemphasised growth as the primary objective of the party. This tradition is not new, and has been a core principle since 邓小平, Deng Xiaoping.

3. Political scientists disagree on the meaning of the term “populist” yet there's convergence in which regimes are classified as populist, so I see this distinction as valid.