Thoughts on economics and liberty

Response from Harold Demsetz to the questions I raised

Harold Demsetz has very kindly responded in detail to to the questions that I raised here. I'm grateful to him for sparing the time to do so. I hope his response will also prompt those who have not yet done so to read his brilliant article. In the age of the internet we are fortunate that ideas of this path-breaking nature can illuminate us all so rapidly. 

In particular, the issues that Demsetz discusses are pivotal in helping demarcate the boundaries of the state from first principles. I therefore believe that his paper has huge implications for political theory, and hence for public policy. (Precisely how discussions about such fundamental issues could translate into public policy – well, Deepak Lal's paper I discussed yesterday is an example.)  

I won't respond right away to Demsetz's comments (as is often my wont!, often in great haste) but will first think through the whole issue ab inito and make sure I have properly understood all relevant issues. To me, having a very clear personal understanding of such matters is important in order to develop a consistent theory of the state that I'm trying to propose, in The Discovery of Freedom

I'd like to encourage you to freely comment on this issue, including providing other examples. Social policy and environmental policy can, I believe, throw up endless examples to test Demsetz's findings. We must ALL understand what Demsetz is saying, for it has very significant implications for the way our governments are structured and what they do. Members of the Freedom Team of India are, in particular, requested to pay great attention to this paper.


Dear Sanjeev:  

1) My article, 'The Problem of Social Cost: What Problem?', which you kindly recommend to your readers, does not claim there are no resource allocation problems in a well-defined competitive private property right system.  It claims that those problems that ascribe inefficiency to positive transaction cost suffer from the logical error of determining efficiency without  recognizing the necessity for treating transaction cost as a relevant part of the calculation of private and social costs.  If my article is correct in this respect, it implies a considerable reduction in the numbers and types of cases that would be properly identified with misallocation of resources. This was my main objective.     

(2) The last part of my paper associates true resource misallocation problems with strategic, free-rider problems, whose essence is not transaction cost but strategic behavior leading to misrepresentation of true benefits and costs (and coupled to a presumed superiority of government to substitute more accurate values for those revealed in private markets).  I explicitly refer to problems of air pollution and global climate as examples of such problems.  It may well be that there are other classes of problems that reflect true inefficiencies but which are not adequately exposed in my article, although certain kinds of problems would not be appropriate, such as those that rest on irrational behavior.  Even if we find additional problems, the validity of my arguments would still imply a substantial reduction in the number of cases now judged to reflect resource misallocation that actually identify true misallocation of resources.   So, lets consider the cases you put forth to see if they ought to be added to those that are strategic in nature.        

(3)  Sanjeev's difficult issues:

    (a) Issue 1 asserts it is impossible to define some resources and to  attach property rights to them.  True enough, but all we learn from this is that there is a high cost to privatizing or even to publicizing.  The cost of privatizing is something that I do discuss, but it is different from transaction cost.  The notion of transaction cost I use is that which is common to the externality literature — a cost of striking agreements between interacting parties.  This cost may well be so high that transactions do not occur.  The difficulty in even defining the objective of the transaction has no clear bearing on the cost of transacting that is incurred to exchange whatever definition, tight or loose, is given to the object.  If the object cannot be defined at all, it is senseless to talk of transaction concerning 'it.' But so what?  It is efficient not to have a transaction, in such a case.   I do not claim there are no problems in this world that we wish we could solve more cheaply, such as defining and measuring an asset; I wish we could produce steel more cheaply, but some steel is just not worth it cost of production.  So what? 

    b1 ) Smokers will not pay.  Sanjeev confuses the issue of second hand smoke by mixing it up with problems of owning the air.   Sanjeev claims that if smokers have the legal right to smoke they won't pay the bartender for damages done to him by their smoking.  This is untrue if the bartender has a right to serve or not to serve, for smokers who value a drink more highly than the price that makes the bartender willing  to bear the risk will pay; others will not, and will not smoke at the bar.   Sanjeev next claims smokers won't pay because they will not be held liable if they smoke at the bar without paying.  He neglects their inability to get a drink if this is the case, but, nonetheless, he goes on to say they won't pay because no court would hold them liable for damages because each smoker causes so little harm that the court would have difficulty measuring  the harm and identifying its source.  What difference is there between this and a little small scale stealing?  Sure, stealing takes place, but this is not an externality problem brought on by transaction cost; its a problem in enforcing ownership rights.  If the bartender has a right to control who does and who does not drink at his bar, the smoker who does not pay the bartender is liable for the amount the bartender posts as the price for smoking at the bar (plus some deterrent fine).   The harm that is or is not done to the bartender's lungs has nothing to with the bartender's exercise of a right to exclude those from drinking who do not pay. 

    b2)Here Sanjeev assumes smokers have a right to smoke while they drink at the bar. I note in my first reply (on Sanjeev's prior blog) that this degenerates into a free rider problem (much like highway congestion) where people who do not really desire to smoke will smoke just to get paid for not smoking.  Since this is a strategic behavior case, it is already covered in my article. He then says that the price the bartender will need to pay will be too high to preclude smoking by true smoking addicts, and it is this price that the bartender will need to pay.  Why? The bartender can ask a lower price and accept the fact that it will eliminate fewer smokers.  Problems of addiction are problems of addiction, not problems of externalities.  A true addict will smoke even if the State makes smoking illegal, given that no penalty is high enough to keep him from smoking.    Well, then, since price cannot deter him, the bartender simply refuses to serve smokers; if the addict persists in smoking but does not win the bartender's permission, he simply gets hauled to jail.  Addiction here is just a substitute for wealth.  That is, a patron is so wealthy that he'll smoke at the bar even if the bartender prohibits smoking; well, then, the smoker gets hauled away for violating the bartender's rights.  And if the bartender did not have a right to keep smokers from the bar, he could never offer a payment larger enough for a would-be smoker to stay away from the bar, whether this is because the smoker is an addict or because he is so wealthy there is no sum the poor bartender can offer that is large enough.   Wealth distribution problems are not externality problems; that is, we can always change the distribution of wealth in ways that manipulate the sums that interacting parties are willing to pay to influence an interaction.  Similarly, we can always change tastes to influence what interacting parties are willing to pay, and creating an addict is equivalent to changing a person's tastes.  — Harold 

Please follow and like us:
Pin Share

View more posts from this author
2 thoughts on “Response from Harold Demsetz to the questions I raised
  1. Sanjeev Sabhlok

    My interim email to Harold Demsetz today (very busy weekend; no time to think, unfortunately). I'm posting this for info of my blog readers, and request them to provide comments, please. The matter of determining the boundaries of the state is crucial. We are obliged to think through all such details and articulate a clear and consistent theory of the state. Your comments will help everyone understand the matters better. So please don't hesitate to ask!


    Dear Harold

    My apologies for the continuing delay in getting back to you on your comments. Given this delay I thought I should at least update you on my interim thoughts, given the significant effort you've put into the discussion. 
    Your comments have shown me that I prematurely considered the government intervention option in this case. However, since I've not had enough time to think though issues, I'm still grappling with a few matters of detail. My theory of state derives from a precise equality between freedom and accountability. The state, in my view, should ensure accountability where voluntary bargains fail to account for serious harm. It is by now clear to me that subject to the risks being disclosed and bargains being freely conducted, a range of voluntarily options could ensure efficiency – and accountability. But what if the necessary conditions are not met? Indeed, what are these necessary conditions, and why? 
    I hope to take time off work soon, to spend time to clarify my mind fully. I'll revert with follow-up comments at that stage. Once again, thanks very much for the time you've taken to interact with me.

Social media & sharing icons powered by UltimatelySocial