Thursday 5 July 2018

Why Kaushik Basu's 'Republic of Beliefs' is mischievous nonsense

Kaushik Basu is a celebrated Indian economist. Thus, anything he writes must not only be utterly foolish but also the most mischievous thing he could write, given his own preferences, under the circumstances.

Thus in 'the Republic of Beliefs' we find this passage-
The first step towards an amended law and economics is to recognize that if the only way that a law has an impact is through its effect on the beliefs of people, its effective conduit has to be the idea of the ‘focal point’, as developed by Schelling (1963). A successful law is one that shifts human behavior by creating a new focal point in the ‘game of life’ (Binmore, 1994) ...
Why is this nonsense?

Coordination games which have focal solutions can be easily policed. If we want to suppress homosexual behaviour, we can do so by policing the type of places which are Schelling focal for homosexual activity. If there are no focal points for an activity, it would be very difficult to police.

Civic Law can arise when either a Stationary Bandit, or a Guild or organization of that sort, takes control of a focal point for trade and gains a reward for contract enforcement. However people will limit their exposure to this jurisdiction by hedging. This does not depend on 'legitimacy' at all- just incentive compatibility.

Basu believes otherwise-
 the law being an instrument of helping the citizenry is an important precondition for it to be focal.
This is quite false. Unless there is already a focal point for a particular activity, it would be pointless to have a law regarding it. What actually happens is that the law asserts jurisdictions over all activities at a particular focal point. If it can generate more revenue for itself, than the cost of enforcement, we say this is an incentive compatible and sustainable arrangement even if no one likes it.
An illegitimate government, foisted on a people to extract rents out of them, will have difficulty in being effective.
The East India Company was very effective. The Mughal Emperor though legitimate and a focal point for mutinous sepoys and various other malcontents was wholly ineffective. The Company was replaced by direct rule. The British never again made the mistake of letting natives have charge of the artillery.
All this does not mean that it is impossible for an exploitative and illegitimate government to be effective. It simply makes the task harder for such a government.
What great difficulty did the British face? Which Viceroy spent sleepless nights? By comparison, elected leaders, post Independence, have found ruling India very difficult.
It may have to split the interests of the players, letting a small minority gain, inflict a large loss on the majority and siphon off the difference. Colonialism is a good example of precisely this.
Why? Because the Colonial powers had superior shipping technology and thus were more fiscally solvent than indigenous rivals for power.
Unlike the ancient conquering kings, colonialism was a fine innovation in management, whereby you use a handful of the local people to exploit a mass of the local population.
The same thing happened under feudalism.
The amazing feature of European colonization in Africa, Asia and Latin America, and especially the former two, is how few of the colonizers were actually needed to run and exploit these large economies.  
This wasn't amazing at all. What mattered was that the Europeans were more fiscally solvent because they handled Oceanic trade.
...people are more likely to adhere to the rule of abiding by the law purely because it is the law and even when it goes against ones narrow self-interest, if they treat the law as legitimate and feel that the state has some legitimacy (McAdams, 2015; Feldman and Teichman, 2009; Bilz and Nadler, 2009).
Nonsense! We all feel, as Bork argued re. Griswold, that no law, however legitimate, needs to be obeyed if we think it silly and, moreover, will get away with breaking it. Bingo may be a form of gambling, but we don't believe a Bingo drive organised by the local Church actually amounts to a violation of laws against gambling.
Interestingly, we can distinguish between two notions of legitimacy, which I shall refer to as first- and second-order legitimacy. First-order legitimacy is a kind of minimal legitimacy whereby the functionaries of the state, that is, those who have taken up jobs where they are supposed to uphold the laws of the state, abide by the law.
Utterly foolish! Inspector Clouseau may go around arresting every one, including the President of the Republic, on the basis of violations of some arcane laws but he does not have first order legitimacy at all. He will be transferred to some rural shithole where the local populace will kick his head in if he tries to act smart.
That is, they enforce the law as they are expected to, but the citizens are driven by self interest and conventional utility maximization. Second-order legitimacy, on the other hand, is defined as something more all-embracing. This is the case where all players—citizens and functionaries alike—are imbued with the value of following the law because it is the law.
Which can never happen because people aint stupid.
In case the law has first-order legitimacy we are in the world of traditional law and economics,
Nonsense! We are in a Pink Panther film with Peter Sellers playing a bumbling cop.
where a new law changes the game citizens play since the functionaries robotically follow the law. On the other hand, if the state and the edicts of the state have second-order legitimacy, the law will not only be effective but it will be so without the need to have anybody enforcing the law.
So, we will be in La La land where pigs fly because the law tells them to do so.
It is arguable that some high-income countries do approach something like first-order legitimacy and that is the reason why the laws are generally effective in these countries.
Kaushik Basu teaches at Cornell. Would he really be shocked to discover that most of his students regularly break laws relating to narcotics and alcohol and public indecency and so forth?
How such legitimacy is achieved is not always clear but evidently this is something nations should aspire to, since it can make the laws more effective and thereby help growth and development.
Utter nonsense!
In brief, legitimacy of the state is not just good for democracy but may facilitate economic efficiency and development.
Basu served the dynasty which had legitimacy because hereditary succession is a possible focal solution to a particular coordination game. But, it is a very bad solution. India should implement anti nepotism laws and also insist on things like Indian citizenship at Birth for its Head of Government (like the US) and minimum educational qualifications for holding Public Office.

There are three main academic reasons why Basu's economics is utterly shite-

1) Basu is ignoring Robert Aumann's work on correlated equilibria

A Law can create a public signal which can become the basis of a correlated equilibrium. When we hear a person is an ex-con, this affects our dealings with him. It is likely that Service providers whose business may suffer if they hire ex-cons will voluntarily adopt screening procedures and advertise the same. This will occur even if no one believes that ex-cons are more likely to offend. Beliefs are irrelevant. What matters is advertising one's capacity for precautionary due diligence.

The existence of a public signal which can affect one's ability to do business leads to hedging of various different types even if we don't believe the public signal is alethic or will be uttered in a reliable manner.  To reduce uncertainty we are likely to invest in 'costly signals' which show us as belonging to the separating aspect of the new correlated equilibrium.

To take an example, if the Central Bank sends a signal that it has a target interest rate known only to itself, then all other agents will be careful to act as if they have already moved to that new correlated equilibrium on the basis of an unknown public signal. This will happen even if no one believes the Central Bank has the nous to calculate the optimal interest rate.

2) Basu does not understand that Schelling focal points are solutions to coordination games which, however, also cause agents to hedge on 'discoordination' games. They are not magical, nor do they necessarily affect Beliefs.
Basu says- 
'The law works, to the extent that it does, by creating focal points in the game of life or the economy game; and, further, this is the only way in which the law affects individual behavior and collective outcomes.
He gives the following example-
 Suppose a group of people land up in a new lawless island. Each has to decide which side of the road he or she drives on. Clearly, if everybody else decides to drive on the left, it is in the interest of the remaining individual to drive on the left. Likewise on the right. In other words, in this society there are two equilibria— everybody drives on the left and everybody drives on the right. 
This is bizarre- everybody drives in the centre of the road- or that portion of it which is safest and least potholed. The smaller vehicle gets out of the way of the bigger one- or the one driven by the crazier nutjob. Yielding is dictated by a 'discoordination' game.

Univocal Rules for yielding will only evolve in a manner that leads to the outcome Basu suggests if there is a punishment mechanism or if the population is homogeneous and there is a high potential loss from collision. There is no rule for yielding when walking down a street- or maybe there is and nobody told me about it- which is why I often end up apologising to pedestrians coming the other way.

Basu can't show that the law creates focal points. What actually happens is that the Law can create a public signal- example 'points' on your driving licence which add up and can lead to its suspension- and this public signal can become the basis of a separating equilibrium- such that, for example, people convicted of drunk driving are sacked and exposed to public humiliation.

3) Basu thinks 'non-focal'  laws can't be implemented whereas history shows they can.

Suppose I am a masochist. Suppose you are a sadist. We can coordinate by joining an S&M club. However the law can suppress that club. You can be prosecuted even if I consented because the law can stipulate that I had no such capacity and thus the act was illegal.
Basu writes- 
 If a law is enacted that tries to direct the entire society to an outcome that is not an equilibrium and hence, not a focal point, then the law is doomed not to get implemented.
This is quite untrue. Society can suppress repugnancy markets and unconscionable contracts even if they represent focal solutions to coordination games. Indeed, any given Jurisdiction is bound to suppress at least one focal solution so as to create a separating equilibrium. This is because it is in our interests to emit 'costly signals' which have a reputational effect.

It makes sense, for a Monty Hall type reason, to change behaviour after a new law is enacted such that a new  Public Signal is emited because Muth Rationality tells us that the correlated equilibrium has changed.

Basu writes-
 The second reason for non-implementation of the law is extremely important and pertains to a generic and somewhat open-ended problem of focal points—namely, that it is not evident what constitutes a focal point for different groups of people.

No. What is focal is that which is, by 'common knowledge,' most salient.  It isn't the place I would prefer or you would prefer. It is the place everybody thinks everybody else would think is most salient even if nobody actually prefers it. This is like a 'Keynesian Beauty Contest'. We don't vote for the girl we think is hottest. We vote for Kaushik Basu because we believe the only sad losers who bother to participate in such contests are Basu himself and people who look like Basu and probably are Basu but get easily confused and identify as notorious sex-pot, Partha Dasgupta. 

The same signal may be picked up by one set of people to be focal and be ignored by another group.

Focal points are not signals. They are things which have salience for the emitters of signals.  There can be public, Aumann type, signals about what is focal and this may promote a better correlated equilibrium. But, if such signals are ignored in a systematic manner then they aren't actually Aumann type at all. There is no correlated equilibrium. 

We know, for instance, that players sharing a common group identity find it easier to coordinate onto a focal point (Habyarimana, Humphreys, Posner and Weinstein, 2007; see, also, Boettke, Coyne and Leeson, 2008). This, in turn, implies that when a group of people have got reasonably used to a focal point, the presentation of a new focal point can actually make the coordination problem worse
If agents are heterogenous, then there will already be arbitrage between coordination and discoordination games in line with Conley & Nielson. However, this will happen anyway because of hedging or so as to capture income effects. Focality solves coordination games but discoordination games- which are like 'separating equilibria'- always arise if only as a matter of 'capacitance diversity'. 

This does not cause any problem for the implementation of the Law. Different groups will use different strategies to stay clear of punishment. It is not the case that policing becomes impossible in a multi-ethnic neighborhood. Smart Police Commissioners do 'Community Policing' by building trust with what Conley & Nielson describe as arbitragers or 'cultural market-makers' who can ensure that the same nomothetic objective is achieved but in a different, ideographic, manner.

Why is Kaushik Basu's book the stupidest possible response to what is happening nowadays in India and America?

The answer is that he is talking as though countries need to have homogeneous beliefs- Trump's America must have WASP values and norms, Modi's India must be purely Hindu- because 'the Republic of Beliefs' is what determines whether a Society can operate under the Rule of Law.

If some RSS type made such a claim he would be pilloried by his own people. Hinduism denies any such idiocy.

It is a different matter that rights should be linked to obligations only under an incentive compatible vinculum juris. However, that has nothing to do with Beliefs. It is simply a matter of pragmatic, defeasible, Mechanism Design whose aim is last mile delivery of good Governance.






No comments: