A common thread on this blog is the thought that free markets are preferable to various forms of state intervention if our aims are, not only to achieve prosperity, but also alleviate poverty. The empirical correlation between robust markets and prosperity is, I think, undeniable, and the burden of proof is on those who propose governmental interference.
Here I take a different tack. Classical liberals should endorse a political system that includes a safety-net for the poor while simultaneously abolishing virtually all other barriers to market entry. This means no more subsidies, no more tariffs, no more licencing of professions, no more burdensome regulations, no more state-run education, no more barriers to immigration, no more unproductive public spending, and no more bloated bureaucracies (you can add an appropriate public-goods proviso.) Call this view sufficientarian liberalism. The view is sufficientarian, not egalitarian: it advocates state redistribution of resources only toward those who cannot provide for themselves.
Sufficientarian liberalism can be philosophically justified. In the Doctrine of Right Kant argues that to sustain the civil condition the state must provide means to those incapable of providing for themselves. But the state cannot legitimately redistribute resources beyond this, because doing so would encroach on people’s protected freedoms. The only legitimate reason for coercion is the establishment and maintenance of the civil condition. That is why the state can punish criminals: the state hinders the freedom of someone, the criminal, who has hindered the freedom of his victim. Now property-less persons cannot act autonomously because they are subject to the permissions and wishes of others. Therefore, the state must provide them with the material means of acting autonomously, as required by the civil condition. If you are charmed by this view, then you have a first-order, ideal justification of the sufficientarian liberal state.
But suppose that Robert Nozick is right and no redistribution, not even to the poor, is justified. In that case we can no longer justify sufficientarian liberalism on first-order principles. However, we can still defend it as a second-best, non–ideal political arrangement. If the Nozickian utopia is unattainable, then classical-liberals’ best strategy might well be to support institutions that frontally address the plight of the poor. Now imagine a society where the only redistributive job of the state was to help the poor. That society would be an immense improvement over the crony-capitalist systems we endure today. If we couple a safety-net with vast deregulation of markets, and we add the fact that freer markets help the poor more than known alternatives, then the classical-liberal has the upper hand, because the defender of the welfare state has lost her main argument for big government. If the poor are provided for, all that remains of the welfare state are subsidies, privileges, rent-seeking, and various other inefficiencies. I doubt honest egalitarians can defend that.
(There is some empirical evidence that countries like Norway or Denmark are closer to this ideal than, for example, the United States or France. See Jason Brennan’s post here. While of course there are many useless regulations in those countries, property and contract are well defined and barriers to market entry are lower than in other places. This might explain why they do better notwithstanding their strong safety net.)
So, the argument for sufficientarian liberalism is an argument in the alternative. If you share Kant’s claim that autonomy requires providing for the poor but prohibits further redistribution, then you would have provided a principled justification for sufficientarian liberalism. Alternatively, if you think Nozick is right in opposing all redistribution, you may still defend sufficientarian liberalism as a second-best political arrangement. This second-best solution is still enormously better than the statu-quo. Of course, my proposal is still not granular enough: it does not define public goods; nor it does specify which of these other regulations should survive; nor does it explain exactly how the redistribution in favor of the poor will be implemented. But the idea is intelligible enough: either sufficientarian liberalism is morally justified or, if it is not, it is the second-best proposal for the real world. As a political strategy, sufficientarian liberalism may well give classical liberals the long-coveted high ground in political debate.