Rodrik & Mukand's theory of liberal democracy

Rodrik & Mukand have been touting a rather bizarre theory of how liberal democracies originate for a couple of years now.

Does it feature any proper economic analysis or is it junk social science? Let us look-

 We provide in this paper a taxonomy of political regimes, distinguishing in particular between electoral and liberal democracy. We take the main distinctive feature of a liberal regime to be the restraints placed on those in power to prevent discrimination against minorities and ensure equal treatment.

A theocracy may have  a Scriptural injunction against such discrimination. A despot- like Fredric the Great- may prevent discrimination against minorities so as to encourage entrepreneurs and skilled artisans to settle in his domain so as to boost the tax base. He may also wish to attract the best soldiers of fortune regardless of religion.

Avowedly Liberal regimes- like America pre 1965- may have highly discriminatory laws and practices. It is probable that our laws are discriminatory in some manner not obvious to us but which will seem so to our descendants.

In the Seventies, some Liberals argued that laws against paedophile were misguided in the same way that laws against homosexuality were misguided. Cohn Bendit wrote about the 'naturalness' of sex with very small children and claimed it would help 'liberate' society from Capitalist Patriarchy.
The infamous Paedophile Information Exchange (which was exactly as evil as it sounds) was affiliated with the National Council for Civil Liberties.
People with Paedophile tendencies constitute a minority. I'd imagine that most of them find ways to continue to obey the law and avoid harm to innocents. It would be wrong to punish a person for his inclinations rather than his actions. Still, there is a slippery slope here. A guy jerking off to kiddie porn is not accorded equal treatment to one pleasuring himself to wet veshti videos of P.Chidambaram though no direct harm arises in either case. Still, the fact is children were harmed during the production of those evil videos whereas Chidambaram deliberately flaunted his shapely haunches so as to give Dalal Street a hard on.

There may be medical treatments which help paedophiles to overcome the impulse to act in a criminal manner. However, these treatments may have side-effects. Suppose a treatment were shown empirically to have no negative effect. Making the treatment mandatory leads to an Expected Pareto improvement (assume the treatment uses up no scarce resources- otherwise we could still speak of a Hicks-Kaldor improvement). Then a Liberal still could not endorse this measure unless our current notion of 'self-ownership' changes such that sexual preference is no longer a type of 'personal property' that must be respected. This is actually quite reasonable.  We know that what constitutes 'empirical evidence' changes dramatically over time. A 'regret minimizing' reflex on our part has us refuse to let 'hard cases make bad law'. But we could equally call this a conservative or religious scruple. It arises from our fate as sentient beings who evolved on an uncertain fitness landscape.

The common sense view is that 'whatever is best administered is best'. Words like Conservatism or Liberalism are strategic rather than purely alethic. What matter is 'effectiveness'

The restraints can be legal or administrative; they can be maintained by constitutional strictures or self-enforcing agreements. What matters is that these checks, which we associate with “civil rights” for short, are effective in practice. Our focus is squarely on these missing restraints – the relative weakness of civil rights – in illiberal electoral democracies.

So- the 'restraints' are a matter of oversight or interessement by other human beings. That's all that is being claimed. If other people are smart and scrupulous and highly effective then we get one result. If not we get its opposite. However, what will motivate 'other people' to be smart and scrupulous and highly effective in restraining those in power? Obviously, they will do so if they are acting in their own self interest. That's what actually happens when minorities gains civil rights. They organised themselves and made strategic alliances and thus obtained a counervailing power. That is the whole story here. Liberalism does not matter. People do. Sometimes they succeed in building a coalition which protects or advances their interests. Sometimes they don't.

Rodrik & Mukand take a different view-

We argue that the failure to protect minority rights is a readily understood consequence of the political logic behind the emergence of democracy.

Either an emergent democracy wants to fuck over a minority and does so to its heart's content till it discovers that this is a silly thing to do- minorities should be milked not murdered- or else it has no such desire and does no such thing.

'Political logic' is about coalition formation and stability. It isn't about some magical power ascribed to any type of code. Why? Codes are defeasible. They can be read any which way.

Rodrik & Mukand, because they believe in the magic of codes, have an impossible problem- viz. how to account for the 'White Magic' of three different codes coming into alignment in a benevolent way. It seems impossible that Sauron and Saruman and Gandalf should suddenly become univocal and benign.

What requires explanation is not the relative paucity of liberal democracy, but its existence – rare as it may be. The surprise is not that few democracies are liberal, but that liberal democracies exist at all. To make our point, we distinguish specifically between three sets of rights: property rights, political rights, and civil rights. We define these as follows: (1) Property rights protect asset holders and investors against expropriation by the state or other groups. (2) Political rights guarantee free and fair electoral contests and allow the winners of such contests to determine policy subject to the constraints established by other rights (when provided). (3) Civil rights ensure equality before the law – i.e. non-discrimination in the provision of public goods such as justice, security, education and health.

Education and health are merit goods not public goods. There can be- indeed, there must be- severe service provision discrimination, at the margin, in the monopsonistic supply of Defence and Justice. None of the rights Rodrik mentions actually guarantee anything. They stipulate a method of redress which involves its own up-front costs which must be balanced against likely benefits.

We classify political regimes according to which (combination) of these rights are provided. In dictatorships, it is only the property rights of the elite that are protected.

No actual dictatorship has this feature. Even in North Korea, the police will catch a guy who steals your picture of the beloved leader. Why? Small thieves turn into big bandits. Big bandits can seize power. Elite property, however, is up for grabs. Look at Saudi Arabia today. The ordinary Saudi isn't going to lift a finger to protect the property of some Prince. It is the Dictator who decides who is elite and he purposely makes a scapegoat of some of them from time to time.

Classical liberal regimes protect property and civil rights, but not necessarily electoral rights.

Utter nonsense! However limited or unlimited the franchise, classical liberal regimes uphold voting rights in the same manner as property rights. Indeed, in Britain, prior to 1832 many electoral rights were property rights because some Boroughs were 'Rotten'. On the other hand, a 'classical liberal regime' may decide to disenfranchise and ethnically cleanse a particular group. The excuse of war, or the threat of war, or even a perceived threat to the common good, has been used in this manner in Britain.

Electoral democracies, which constitute the majority of present-day democracies, protect property and political rights, but not civil rights.

If a country is a democracy, the minority has a vote. Chances are they will use that vote strategically so it has a high Shapley value. In any case, if the majority is rational, it will want the minority to be more productive and pay more tax to finance public goods and social capital formation.

Liberal democracies protect all three sets of rights. Note that we operationalize the non-discrimination constraint under liberalism as equal treatment by the state in public goods provision in different domains – legal, religious, educational, etc.

Rodrik & Mukand think legal and educational services are non rival. Why not assume they are non excludable as well? Why stop there? Why not say law and education use up no scarce resources? Why do they mention 'public good provision' in the religious domain? Does the US Senate conduct a secret sacrifice to Chthulu to prevent it from devouring the world?

Each one of these rights has a clear, identifiable beneficiary. Property rights benefit primarily the wealthy, propertied elite.

Nonsense. A wealthy propertied elite can hire gun-men to protect their stuff. Poor and cowardly blokes like have to depend on the Police.

Political rights benefit the majority – the organized masses and popular forces.

Not necessarily.

And civil rights benefit those who are normally excluded from the spoils of privilege or power – ethnic, religious, geographic, or ideological minorities.

Any justiciable right favours those with the money and education to appeal for their enforcement. Those wholly 'excluded from the spoils of power' can do nothing. Take Eddie Mabo- the Australian aborigine whose campaign won back some land rights for his people. But for all his courage and eloquence, he could do nothing on  his own. Some big law firm took the case pro bono. The indigenous Australians at that time simply didn't have the money or the connections to get a hearing even though, by then, they did have 'civil rights'.

When the propertied elite can rule on their own they establish an autocracy that protects their (property) rights and little else.

But, if they do so, the value of their own property will be very much less. Why? The workers have no incentive to work hard and accumulate possessions.
In any case, each member of such an elite would have to maintain his own armed retinue to protect himself from the depredations of either disaffected 'social bandits' or else his own peers.
What Rodrik & Mukand are suggesting is 'incentive incompatible' and irrational.

This has been the usual outcome throughout the long arch of history. Mass democracy, on the other hand, requires the emergence of organized popular groups that can challenge the power of the elites.

This isn't true at all. In 1848, there were popular movements in Germany and Britain which failed to achieve anything at all. Why? General Napier explained everything when he pointed out to the 'Physical Force' Chartists that he had canons, they didn't.
The ruthlessness of the elite was plain for all to see. The wealthy could retreat to their castles and unloose vicious mercenaries on the masses. The rich might lose money- but they could make it back. The poor would be killed.

In the 19th and 20th centuries, processes such as industrialization, world wars, and de-colonization led to the mobilization of such groups. Democracy, when it arose, was typically the result of a quid pro quo between the elites and the mobilized masses. The elites acceded to the masses’ demands that the franchise be extended (usually) to all males regardless of property qualifications. In return, the newly enfranchised groups accepted limits on their ability to expropriate property holders. In short, electoral rights were exchanged for property rights.

Parliament can change property rights. In the short run there may be a Legal check, but electoral democracies can change the Law and amend the Constitution. No deal of the sort described above is possible. Why? Even if the leader of the plebeians is an honourable man, there is no guaranteed, under democracy, that he will remain the leader. Someone else may get in and do as he wishes.

What was possible was for the elite to have embraced a 'Corporatist' solution instead of Electoral Democracy.

The defining characteristic of this political settlement is that it excludes the main beneficiary of civil rights – the dispossessed minorities – from the bargaining table. These minorities have neither resources (like the elite) nor numbers (like the majority) behind them. So they do not have something to bring to the table, and cannot make any credible threats. The political logic of democratization dictates the provision of property and political rights, but not civil rights.

Why did the Jews- who were by no means popular- get civil rights during the long nineteenth century? In Tzarist Russia they were scapegoated as they would later be scapegoated by the Nazis.
One reason is that Jews were superb politicians and publicists- almost always of an enlightened sort.

As we formalize in section III, the provision of civil rights is costly to the majority and largely unnecessary for the elite (who can pay for their own collective goods by extracting a surplus from the masses). Therefore the political settlement is one that favors electoral democracy over liberal democracy

Actually, civil rights do matter for the elite. The Earl of Rosebury wants to be able to marry a Rothschild in the knowledge that their son will one day inherit his seat in the Lords; the Junker general who repairs his fortunes by marrying a Jewish heiress hoped (but failed to secure) the same thing ;the Australians were up in arms when the daughter of a prominent Sydney Solicitor married the Rajah of Puddukotai but was not recognised as a 'serene Highness' by the King Emperor.

In business, the desire to exploit economies of scale and scope militate against discrimination and in favour of civil rights. McDonalds doesn't want to have two counters- one for Whites and one for Coloureds. Alfred Sloan, at GM, turned its fortunes around during the Depression by selling cars to African Americans. Later on Detroit came to depend on Black labour from the South.

What is true of the elite is even more true for the poor.  People gain by having a bigger choice set  even in purely personal matters- like marriage or friendship.

Why do Mukand & Rodrik think 'provision of civil rights is costly to the majority'? Mukand is of Indian origin. All Indians have benefited from improved rights for Dalits.  As for America, did Whites really lose from the ending of Jim Crow?

Perhaps they are referring to affirmative action. I suppose that might be resented. But affirmative action is a different kettle of fish from civil rights.

By distinguishing explicitly among three groups and three associated sets of rights, our framework helps explain why liberal democracy is such a rare beast. But liberal democracies do exist, and the question is how they can ever be sustained in equilibrium. We discuss several circumstances that can mitigate the bias against civil rights in democracies. 

First, there may not be a clear, identifiable cleavage – ethnic, religious, or otherwise – that divides the majority from the minority. In highly homogenous societies, the “majority” derives few benefits from excluding the “minority” from public goods and suffers few costs from providing equal access. This may account for the emergence of liberal democracy in Sweden during the early part of the 20th century or in Japan and South Korea more recently.

The theory of price or service provision discrimination tells us that the monopsonist (e.g. the Govt which supplies legal and police services) can segment the market if the barrier corresponds to a signal which it is easy to detect but costly to evade. It is difficult for me to change my colour or gender or to pass myself off as a child. It is easy to identify such signals. In a homogeneous society, only gender and age are easy to detect once a certain material threshold has been passed. Colour or Religion or Class would be easy to disguise and costly to accurately determine.

Japan, since the feudal period, has kept records of a person's class origin. This is why 'untouchability' persists.  It is routine in Japan, but no where else, to employ a private detective to determine the other party's lineage before marriage. Sweden has a very different history. Though it had a proud Aristocracy, it chose a French King who was quite humbly born. It's political development is a tribute to its own indigenous virtues and values.

What accounts for Japan and South Korea's (more recent) 'liberal Democracy' is American involvement. But Abe is an aristocrat and the last South Korean leader was the daughter of a dictator.

Second, the two cleavages that distinguish the majority from the minority and the elite from the non-elite may be in close alignment. In such a case, the elite will seek both property and civil rights as part of the political settlement with the majority. Think, for example, of the position of the white minority government in South Africa prior to the transition to democracy in 1994.

South Africa is a poor example. There has been a huge 'White flight'. It is by no means certain that South Africa won't go the way of Mugabe's Zimbabwe which, by the way, was supposed to have made the same sort of 'deal' and thus been established as a liberal democracy.

The truth is- whether we are speaking of Burma or Idi Amin's Uganda or Mugabe's Zimbabwe- denying property or civil rights to a minority is the first step towards denying those rights to everybody. Rights are interdependent. Taxation without Representation means either fiscal collapse (absent a 'resource curse') or endless rent contestation. Civil rights and property rights are connected by concept of fungibility. If you can't sell or mortgage a given asset to a specific class of agents, then fungibility is affected. An arbitrage opportunity is created in a manner adverse to public good provision.

Third, the majority may be slender and need the support of the minority to mount a serious challenge to the elite.

In which case the elite will buy off the minority to side with it. After all, it is the elite which has the cash.

Or there may be no clear-cut majority, with society characterized by a preponderance of cross-cutting cleavages.

Politics is about the forging of identity classes. Sooner or later a political entrepreneur will find some mimetic 'signal' of the required type.

In these cases, repeated game incentives may ensure that each group recognizes the rights of others in return for its rights being protected by them. Lebanon’s “consociational” democracy may have been an example of this, before differential population growth and outside intervention upset the pre-existing balance of power among different religious denominations.

Lebanon was not an example of 'repeated game incentives'. It was purposely set up by the French as a sectarian state which the Maronites were supposed to dominate. The thing was wholly unsustainable. There was no 'pre-existing balance of power'.

So far I have focused on Rodrik & Mukand's ignorance of history. Let us now look at whether they know Economic theory.

Why Rodrik & Mukand think civil rights impose a cost.

We label the three groups in society with the subscript 𝑖, with 𝑖 taking one of the three possible values e (elite), a (majority), and b (minority). Members of each group derive utility from their (after-tax) income 𝑦𝑖 and from consuming a public good 𝜋𝑖. (1) 𝑢𝑖 = 𝑦𝑖 + 𝜋𝑖. We normalize the economy’s total output to 1, with the pre-tax/transfer shares of the elite and non-elite given by 𝛼 and (1-𝛼), respectively. Total population is assumed to equal a mass of 1+ ε, where the elite constitute a minority ε of the population but control more than half of pre-tax/transfer output (𝛼 > 1 /2 ). The non-elite have mass of 1 and are split between a majority and a minority, with population shares n and (1-n), respectively (n > 1/2 ). 

Either the elite have more than half of output without any public good provision or they need a specific public good to be sure of getting it. Suppose they use the minority exclusively to get it- i.e. only members of the minority are employed in the police and judiciary. Then the majority is only worth doing a deal with if either
1) there is an expected welfare gain for the elite by doing such a deal (perhaps it raises the productivity of the majority which yields a capital gain on the elite's assets). In that case, Elite utility from a specific public good  will be a function of the expected increment in elite income by doing the deal. Notice that utility and income are no longer independent.

2) there is a sufficiently substantial deadweight loss associated with the current arrangement whereby the elite use the minority to shear the majority like sheep. In this case, some underlying mechanism is crashing- otherwise why is the deadweight loss now suddenly so substantial? Notice, this means that something important is happening in the economy which this system of equations don't capture.  Thus the model is under-specified.

In the absence of any taxes or transfers, 𝑦𝑒 = 𝛼 and 𝑦𝑎 = 𝑦𝑏 = (1 − 𝛼).

WTF? How is the income of the majority equal to the income of the minority? This is crazy stuff! Not only is the model worthless but it is also utterly mad! Suppose the rich employ ten percent of the population to enslave the remainder. This 'Mameluke' caste may be of a different, martial, religion or geographic area, or distinctive ethnicity. It would be a very strange coincidence if their share of income were exactly equal to that of the enslaved majority.

The correct equation for the above is that ya +yb= 1- alpha.

The gap between 𝛼 and 1/ 2 is a measure of the class (income) cleavage.

Nonsense! What matters is income differentials- e.g. minority gets twice the wage of the majority but the elite get a hundred times that, on average.

We model the identity cleavage by assuming groups exhibit differences in the type of public good they prefer. The type of public goods is indexed by 𝜃 ∈ [0,1]. The three groups’ ideal types are given by 𝜃𝑖, 𝑖 ∈ [𝑒, 𝑎, 𝑏]. The utility derived from the public good thus depends both on the aggregate expenditure on it and on the type of public good that is provided. There is a deadweight loss associated with the provision of public goods, which increases with the level of expenditures and the gap (from the perspective of each group) between the type that is provided and the preferred type. Denoting total expenditure on the public good by 𝑟, the utility derived from the public good is thus expressed as follows: 𝜋𝑖 = 𝑟 − {1 + |𝜃𝑖 − 𝜃|} 𝛾 /2 𝑟*r, where 𝛾 parameterizes the magnitude of the deadweight loss relative to the direct benefits associated with public goods provision. Note that deadweight loss is minimized, but not eliminated, when 𝜃 = 𝜃𝑖. We shall normalize the majority’s preferred public good by taking 𝜃𝑎 = 1.

There are circumstances where the majority's favoured 'public good' is 'kill elites and enslave minorities'. Doing the killing may itself provide utility. The equation given above is false. Utility from a public good does not depend merely on expenditure on it less the deadweight loss caused by the tax which finances it.
The only situation where the equation would have salience is if it could arise in a non-coercive repeated game with common knowledge etc. But, unless 'elite', 'majority' and 'minority' correspond to different productivity types there would be no income inequality in the first place! But, in that case, the use of these words is misleading.

My conclusion is Mukand & Rodrik are ignorant of economics. This is 'junk social science'. What motivated their spewing up this drivel?
The answer is that they wittingly or unwittingly rely upon racist and sexist and homophobic narratives which historians have long disproved.

Notably
1) British liberalism granted 'civil rights' because....urm... the Brits are just different okay?

The fact that early liberals in the West were in large part the wealthy property-owning elite led to the bundling, in the minds of subsequent analysts, of two kinds of distinct rights: property rights and civil rights. This peculiar, and peculiarly British, history does not fit the experience of other, especially non-Western countries very well. In particular, the elite would often turn out to be interested primarily in property rights. Civil rights were for others, chiefly ethnic, religious, or other minorities.

WTF! Ottoman Turkey gave better civil rights to women and gays at an earlier date than Britain.

It wasn't bleeding heart Liberals who gave civil rights to Catholics- it was the true blue Duke of Wellington & his brother who pushed that through so as to avoid a costly Irish campaign.

What about 'the right to association'? Is that not a Civil Right? Do Mukand & Rodrik really not know that British Whigs fought tooth and nail to deny this fundamental right to Working Men? It is ridiculous to suggest they 'bundled property and civil rights' in their minds.


2) South Korea is different from North Korea not because of who its Super Power protector is but...urm...South Koreans are just different okay?

we would argue that Korea’s liberal democracy has much to do with the relative absence of identity cleavages and the leading role played by the labor movement in mobilizing against the military/industrial elites.

Why did India- with big identity cleavages- remain a democracy wile Korea yo-yoed? D'uh! It was the existential nature of the military threat they faced. There was always a strong labour movement in Korea. Later, during the Gwangju Uprising, students too played a vanguard role. So what? They could be crushed. Soldiers would obey orders to shoot because they knew that if the power of the Army crumbled then there was a good chance the Commies would ultimately take over in which case they would be tortured into signing confessions before being granted the mercy of a bullet in the brain. Their kids, if they survived, would be a permanent slave class of a type familiar from Korean history. The Chinese change in direction and the collapse of the Soviet Union changed the horizon. Even now, we see the Koreans re-pivot as their Northern neighbour's strategic horizon has changed. But then the South Koreans have always known they have to pull their weight- which is why their boys did so well in Vietnam- and, ultimately, sort things out for themselves. Liberal Democracy is part of this delicate dance. Kim the Fat was educated in Switzerland. He knows what is being offered by the new regime. In the end, it seems a piece of pure luck that Park's daughter fell when and in the manner that she did. But, Koreans make their own luck and do it as well as they make the OLED screen on which I am reading what I've typed right now.

3) Lebanon, prior to the civil war, which lasted from 1975 to 1990, it was a model democracy because...urm...well, it had a lot of Christians and Christians are just nicer okay?

Why did this 'model democracy' need US military intervention in 1958? What sort of model democracy is openly sectarian with the top job reserved for people of one particular religion?

4) South Africa is a liberal democracy because....urm... the financial markets just think its different okay?
Apartheid fell because markets thought its debt was junk. This meant that there was an expected increment in income for a new elite coalition from a particular regime. Now things have come full circle. China will decide what happens next. Rights don't matter. Unless of course domestic financial markets are deep in a certain special sense. But that would only be true if the rights' regime is near optimal in some endogenous sense.

Rodrik & Mukand end with some hilarious value judgments. Thus Greece and India don't have civil rights while the Czechs- who sterilize Romani women and criminalise their men- get classed as 'liberal'. Why? Would Modi or Tsipras be willing or able to deny tenure to a Professor for taking part in a Gay Pride march?