David R. Henderson  

Hayek on Social Justice

Magna Carta Club, Part 2... M3 Plummeting...

Yesterday I was reading excerpts from Hayek's Law, Legislation, and Liberty to prepare for a conference this coming weekend in Seattle. Here are some choice excerpts from Hayek's discussion of "social justice."

For the same reason that the prices which guide the direction of different efforts reflect events which the producer does not know, the return from his efforts will be frequently be different from what he expected, and must be so if they are to guide production appropriately. The remunerations which the market determines are, as it were, not functionally related with what people have done, but only with what they ought to do. They are incentives which as a rule guide people to success, but will produce a viable order only because they often disappoint the expectations they have caused when relevant circumstances unexpectedly changed. It is one of the chief tasks of competition to show which plans are false. [italics his]
It's clear from context that "ought" here is not used in the usual sense of "moral obligation," but in the sense of what one should do if one wants the highest possible return on his (her) resources.


Men can be allowed to act on their own knowledge and for their own purposes only if the reward they obtain is dependent in part on circumstances which they can neither control nor foresee. And if they are to be allowed to be guided in their actions by their own moral beliefs, it cannot also be morally required that the aggregate effects of their respective actions on the different people should correspond to some ideal of distributive justice. In this sense freedom is inseparable from rewards which often have no connection with merit and are therefore felt to be unjust. [italics his]

In explaining how it's good to allow new products on the market even though doing so hurts current producers, Hayek writes:
Of course, those who as a result will be deprived of their former customers will incur a loss which it would be in their interest to prevent. But like all others, they will have been profiting all the time from the repercussions of thousands of similar changes elsewhere which release resources for a better supply of the market.
The known and concentrated harm to those who lose part or all of the customary source of income must, in other words, not be allowed to count against the diffused (and, from the point of view of policy, usually unknown and therefore indiscriminate) benefits to many. We shall see that the universal tendency of politics is to give preferential consideration to few strong and therefore conspicuous effects over the numerous small and therefore neglected ones . . .

Comments and Sharing

COMMENTS (10 to date)
david writes:

Interpersonal utility comparison fail by Hayek, clearly...

Non-Austrian neoclassical economists get around the problem of new innovations hurting existing producers by reformulating the problem so that the new innovators compensate the existing producers for their loss; if the new innovation is really better, then it would have enough income to perform said compensation, yes? Then we have a Pareto improvement.

kebko writes:


Amazon.com clearly out-innovated overpriced-stuff-for-sale-with-a-poor-interface.com. When the lost firm was at its peak, it was worth $20 bazillion.
So would we compensate internet investors for the firms they owned in the year 2000? Would that be a Pareto improvement?

Yancey Ward writes:
reformulating the problem so that the new innovators compensate the existing producers for their loss

In what way is that reformulation of the problem?

Bill writes:
Interpersonal utility comparison fail by Hayek, clearly...

Are you being facetious? It's hard to tell.

david writes:

No, I'm not being facetious. Austrians and neoclassicals split over more or less this issue; both agreed that we couldn't measure cardinal utility. The Austrian school gave up on mathematical utility formalization altogether, talking of value scales and deduction only from axioms. The neoclassicists hung on to the math for another half a century and eventually came up with a formalized ordinal-utility foundation that avoided the problems of cardinal utility.

And here we see why the neoclassicals won the fight; when you abandon math, you invariably make all sorts of careless errors like the one Hayek makes here. He concedes that concentrated harm happens; on what possible basis does he assert that this harm is smaller than than the added small benefits to the many? Isn't this comparison implicitly cardinal, albeit buried under a lot of verbiage to hide the error?

(and the neoclassical solution is, as before, to attach compensation to the innovation so that the net change is Pareto improving; if it really is beneficial on net, then it will be able to fund said compensation).

Bill writes:
(and the neoclassical solution is, as before, to attach compensation to the innovation so that the net change is Pareto improving; if it really is beneficial on net, then it will be able to fund said compensation).

And you believe that is workable?

So many questions:

Who gets and pays the cardinal compensation under your ordinal solution? Is this monetary or is it some kind of Harrison Bergeron shackling of the innovator?

Innovations rarely happen in a vacuum. Realistically there are several underlying platforms and innovations that have contributed to this higher innovation (use the Amazon.com example). How do you determine the relative pain amongst numerous players? And then further, the players that contributed innovations under the platforms underneath the higher innovation?

How many variables are you considering in this Pareto improving? Are you considering all the decisions the various recipients of the diffuse benefits make with their new found wealth?

I don't think Hayek is saying you can make a cardinal comparison of net benefits. I think he is saying that the situation is far too complex for you to even make an ordinal determination that includes all the significant players.

You have written nothing to dissuade me from agreeing with Hayek.

Yancey Ward writes:


Hayek himself couldn't have written that better.

In his speech on The Quest for Cosmic Justice, Thomas Sowell selects a good Hayek Quote:

The late Nobel Prize–winning economist and free-market champion Friedrich Hayek, for example, declared, "the manner in which the benefits and burdens are apportioned by the market mechanism would in many instances have to be regarded as very unjust if it were the result of a deliberate allocation to particular people." The only reason he did not regard it as unjust was because "the particulars of a spontaneous order cannot be just or unjust." The absence of personal intention in a spontaneous order—a cosmos, as Hayek defined it—means an absence of either justice or injustice. "Nature can be neither just nor unjust," he said. "Only if we mean to blame a personal creator does it make sense to describe it as unjust that somebody has been born with a physical defect, or been stricken with a disease, or has suffered the loss of a loved one."

Tracy W writes:

Dave - I think the evidence Hayek was drawing on about the benefits of innovation outweigh the costs is that standards of living have risen in Western countries, and that the mass of immigration tends to be to Western countries.

ionides writes:

Forcing innovators to compensate the losers adds to the cost of innovation, reducing the incentive to engage in it. Total resources devoted to innovation and therefore total innovation will fall. Resources will be diverted into lost-rent-seeking activities; a new industry of identifying the loser will emerge. Furthermore, compensation will reduce the incentive of obsolete produces to liquidate resources so as to make them available for profitable employment elsewhere; it would make sense to continue in loss-making behavior to substantiate the claim of harm done. Most of the Pareto improvement will be squandered by this activity. The compensation test is a principle of measurement, not a program for action.

Comments for this entry have been closed
Return to top