Thursday, March 30, 2006

French Unemployment - Some Conjectural Responses to Prof. Romer

1. Students across France have been protesting a new law that would make it easier for French employers to fire young workers. How might the passage of such a law affect the unemployment rate and the labor force participation rate? Why?

ANS: The passage of the law would increase unemployment rate because, as more young people are likely to fired the labor force participation would increase - as previously discouraged workers might fancy a chance of finding a job which was earlier filled by one of their contemporaries. So job search increases --> LFPR increases --> unemployment increases. But this is not a bad thing -- a more dynamic labor market leads to a higher probability of better fits between employer and employee that leads to higher output.

2. France has an unusually high minimum wage. Consider two groups: young people who have completed high school and drop outs who haven’t. Which group would you expect to have more unemployment because of the minimum wage? Would you expect this unemployment to be short-term or long-term?

ANS: A minimum wage results in firms internalizing a productivity threshold, upon crossing which it hires newer people. Arguably, high school completed individuals have "higher" productivity than drop outs. So, a minimum wage might result in the lower quality individuals being "non employable" given the productivity standards that firms have. Of course, I assume rather naively, that wages are equal to the marginal product of their respective labor. In presence of continued minimum wages, one might observe a remarkable divergence in productivity between high school educated and drop outs - given the ex-ante fact of high school educated get employed and learn on the job; while the drop outs end up with dynamic decline in productivity over time.

(3) Suppose the poor labor market for French youth encourages French students to stay in school longer than they would otherwise choose to, earning postgraduate degrees. What long-term effect would this have on the French economy?

ANS: All else equal, the failure to enter into the labor force results in lowered output and lowered per capita income. However, an alternate scenario is that increased post-graduate degrees and students might result in higher levels of innovation and, inevitably, a transition in the economy towards skill-biased production mechanism & less labor intensive. On the whole, a cleavage in the society between the "skilled" and the "unskilled" seems inevitable.

Monday, March 20, 2006

Indentured Labor and Indian Economic Development

It is widely acknowledged that the biggest problems vis-a-vis the functioning of the Indian state - in terms of taxation, jurisdictional capacities and provision of public goods -- in the future comes from how the BIMARU states evolve/devolve relative to the others. (BIMARU = Bihar + Madhya Pradesh + Rajasthan + Uttar Pradesh. BIMARU is a sly abbreviation which means, literally, "sickly" in Hindi/Awadhi/Maithili etc.). (Note, the BIMARU description subsumes the newly founded states of Jharkand, Uttaranchal, Chattisgarh. It also doesn't include Orissa - perhaps the most blighted of all the above states.) Travelling in the second-class compartment in the interiors in any of these states reveals the institutional collapse, the abysmal human development index and all in all a pervasive sense of deprivation.

(As an aside, just finished reading "The Elusive Quest for Growth" by Professor William Easterly - after reading Professor Sen's B+/A- review of Easterly's new book in the Foreign Affairs. Easterly writes of something macabre called "Rawlings Necklace" - to a milder extent, these states form a bureaucratic equivalent of the Rawlings Necklace.)

To a very large extent, the problems that lie there are due to (a) systematic institutional malfeasance and under-investment in human capital and over-investment in strategic, and expedient, white-elephant projects that follow the political cycle (b) an overabundance of natural resources that led to a quid-pro-quo nexus between the Delhi based technocrats and the local elites who allowed minerals/resource extraction in so far as the local power structures were not disturbed. Most of this analysis is well known and largely agreed upon to varying degrees from the Jaswant Singh led Right to the Prakash Karat led Left.

My conjecture is that there is a deeper historical problem in terms of institutional quality in these parts of India -- that cannot be just limited to the last 50 years. An obvious start in this regard is the primary historical evidence of this presented in the difference-in-difference based estimators by Professors Banerjee & Iyer who investigate the impact of property rights regimes on subsequent public investments. In essence, they find that extractive property rights regimes in certain areas results in systematic lowered public goods investment. To my best knowledge, there have not been other such rigorously done empirical studies that study Indian institutional quality and their effect on the concommitant economic growth.

Following is possibly an alternative test, along the lines of the fun and insightful paper by Professor Nathan Nunn of the University of British Columbia. Many of the indentured laborers taken from India to Fiji, South Africa, Mauritius, the Carribean etc were from (a) the economically backward classes (note caste distinction is unclear) (b) they were struggling under the yoke high usury and exorbitant credit market rates (c) the domestic agricultural lands were out of their hands. Data on the number of migrants extracted from India over a period of one-and-half centuries is available in libraries across Australia - as far as I know.

The effect of institutional quality (IQ) on economic growth (G) is typically tested in the
G_{it} = \alpha + IQ_{it} + \epsilon_{it}
framework. However the Corr(\epsilon_{it}, IQ_{it}) is not equal to zero. i.e., the unaccounted factors of growth, subsumed in \epsilon_{it} are most likely driving IQ_{it}.

So, one needs a valid instrument. A valid instrument for Institutional Quality must fulfill two criterion. It must not be correlated with the epsilon terms (i.e., not correlated with drivers of present economic circumstances), and it must be correlated with institutional quality. My sense is that, and aided by Nunn's claim and demonstration, is that locales of India where indentured labor was extracted were precisely those where extractive institutions were already in place. I think, a valid instrument for institutional quality is thus number of indentured laborers extracted over the 1700s & 1800s & early 1900s as an instrument for institutional quality.

I do not know what the results might be- but have a suspicion that the results might validate the Banerjee-Iyer results.

Thursday, June 02, 2005

What's in A Name: Taxation, Warfare and Class Struggle!

The fifth richest man in the world, reports Forbes, is Alwaleed bin Talal bin Abdul Aziz al Saud- a name which literally reads: Alwaleed son of Talal who is the son of Abdul Aziz who belongs to the Saud clan. Or consider the name of the writer of Don Quixote, Miguel de Cervantes y Saavedra. In the Anglo-Saxon world we find typically today names that are truncations of latinate derivations. (In PULP FICTION by Quentin Tarantino, in one of the scenes where Bruce Willis' character kills a boxer in a knockout game - the taxi driver, named Esmeralda, asks him, "what is your name". Willis grunts and says, "Butch". She retorts, "What does it mean?" With his trademark smirk WILLIS replies, "Our names don't mean shit these days.")

Casual empiricism leaves one to conclude that across humanity there is a wide variation in how individuals are identified. That is, there is a question of why do certain cohorts (identified usually on a spatial dimension) have longer names than others. A more anthropological question is, are names solely bonds across generations as in HRH Alwaleed's case; and if so, why does that practise die out say in the West? Two harder questions are: (1) Are naming strategies optimal reactions to changes in underlying socio-economic conditions? (2) Empirically, can variation in naming strategies be decomposed into within group and between group variations.

(To wit, A personal anecdote: In India, there used to be names like Mohandas, Jawaharlal, Subramanian etc etc. These days, it seems (I am most likely wrong) every second male child born is a Rahul or Raj or Karan, every girl is a Ria or Dia or Piya... Am I mistaken???)

Consider the following conjectures.
(1) In preliterate societies, where population is typically low - i.e., and population mobility is very limited the name typically has little relevance in terms of network hierarchy. As population sizes increase, there emerges a pecking order and a "Big Man" emerges. ("Big Man" in modern development economics is a pejorative; see Robert Kaplan's "The Ends of the Earth: From Togo to Turkmenistan, from Iran to Cambodia, a Journey to the Frontiers of Anarchy " to read a visceral description of the catastrophe the "Big Men" induce in modern societies.)There continues to be a great premium on claiming relationship with the elite few in an agricultural society. Names come to then include markers that signals where in the pecking order individual belongs to.

However, if a name is considered as a technology to record the past; a more specific and elaborate name is typically more costly for an individual who belongs to the group that was originally in the low-wealth (thus low-human capital) group in the agricultural regime. If wealth, or access to wealth is proxied by name size, then a caste-intensive society like India it is perhaps not surprising that in equilibrium -- high caste individuals typically have longer and more descriptive names. For example, see the Tamil Brahmins. In contrast, for a society that also has a long continuous historical tradition; but has not developed a critical agriculture induced size; (eg., Aboriginals in Australia) names as identification mechanisms over generations serve little purpose.

(2) Consider, alternatively a situation where tracking an individual involves a cost to the searcher and all individuals are symmetrical and identical on all dimensions other than their names. When the size of the population (and thus the political entity (the State)) increases; coupled with two key factors (1) mobility of population (2) specialization; the identification strategy that every individual ought to find in equilibrium is to adopt longer names that distinguishes them either on basis of family, guild or employment based markers. Not surprisingly, Francis 1 (1494-1547), the king of France issued royal orders that all his subjects take up a surname. Similarly in Henry VIII issued imperial edicts to the same effect. What is perhaps not surprising, is that the two were contemporaries and both France and England were at comparable stages of economic development. However, when an individual's talent are an outlier on one dimension; like Leonardo or Dante or Gretzky a single name suffices to transmit relevant information efficiently through a society's cultural meme. In societies where warfare is incessant between two literate, agrarian and religiously heterogeneous societies; names become a memory device which prevents oneself from homogenization. In Spain, where the Muslims and Spaniards jostled; or Arab tribes that fought amongst each other -- names are inevitably become long and elaborate. (See Jacques Barzun's "From Dawn to Decadence: 1500 to the Present. 500 Years of Western Cultural Life. pg. 113)

(3) If the technology to identify individuals, despite mobility and burgeoning populations, is such that it enables tracking effectively either by individuals or the State with barely any cost. Then name lengths inevitably shrinks. This, I think, has happened in the West -- while in places Multan and Tanjore you end up find babies named Jalaludeen Muhammad Akbar and Srinivasa Venkata Ramanujan.

(Rushdie mentions G.V. Desani's translation of Mahatma Gandhi's name: Action-Slave Fascination-Moon Grocer)

Friday, May 20, 2005

Chronicle of a Catastrophe Foretold?

In a recent blog by Professor Gary Becker and Judge Richard Posner of University of Chicago; they debated regarding the idea of an "optimal" country size given reduced tariffs, diminished telecommunication costs and increased identity consciousness not withstanding this wave of globalization. The idea that political entities have an optimal size in the theoretical literature goes at least as back as Tibor Scitovsky (1958) and Robert Mundell (1968) in lieu of currency areas. But, the true political-economy predecessors of this tradition are the British. The colonial writings of Lord Curzon regarding (unified) Indian Bengal around 1900, the scramble for Africa (no, not Bono versus Bob Geldof) in which the colonial thrust-and-parry games (called Berlin Conference) in 1884-6 etc reveal the modern origins of this idea. The underlying arguments revolved around the tradeoff between administrative gains/costs of running a large entity.

Yet, ironically in the recent Iraq imbroglio few seem to focus on the importance of this issue anymore, barring for the early rumblings of possible tri-vorce of state along ethnic lines - between the Shias, the Sunnis and the Kurds. This idea met understandably with a deaf ear owning to many resonances. Increasingly however, it seems this idea will reemerge owning to the two fundamental inconsistencies inherent in the idea
of Iraq. On one hand the planners and visionaries behind the Iyad Allawi-Ibrahim al Jaafri government insist on maintaining the geographical unit of Iraq as in Saddam's times; on the other hand the exigencies of the transition from an oligarchic to a democratic state-of-the-world forces upon greater instability and thus costs, than perhaps previously anticipated. Consider the following.

(1)Continued prevalence of Iraq as a single unit, implicitly assumes that there are economic and transactionary gain from the original Iraq. That is, internal trade, urban-rural (and North-South) migration and currency units aid in the stability and effectiveness of the nation-state It also assumes that these
gains supersedes the losses due to the inefficiencies imposed in terms of administrative logjams, deadweight losses due to centralized planning processes underway and the continued erosion of the tribal (local) administrative-judicial system owning to increased efforts at 'democratization'.

(2)Further, Iraq typically has been an oligarchic state, where the Baath party officials had sufficient self interest to maintain order, property rights and thus enable low emphasis on redistribtutive taxes. Of course the Iran-Iraq war, the Gulf War, the sanctions etc lay heavy burden on the state machinery; but the common rule of law was maintained with a certain degree of consent, coercion and social sanctions. In contrast, the gains from the democratization process itself presumes that there is free entry and exit for firms (and individuals, ideas and ideologues) and thus enables no rent seeking institutional behavior. By going after the Baath party officials and disbanding the army at first, odious as they are, what has happened it seems that the class of individuals who were most interested in maintaining and fostering property rights (and status quo wealth) have been wiped out. This paradoxically has weakened any immediate gains from the democratization process, at least in the intermediate economic terms. Democracy in absence of property rights leads to old fashioned tribalism; where although decisions are 'democratic' the associated wealth of the citizenry is low owning to free riding and reduced individual effort levels; and more dangerously a difficulty in distinguishing between public and private economic spheres.

So, combining (1) and (2) leads one to conclude that holding onto Iraq as a single geographical unit, has lead to increased economic costs per transaction. An aside, might clarify the issue further. Knowing that bribing only once is enough is often the key to stable economic functioning of the societies. In Kaushik Basu's {Prelude to Political Economy}, he writes about the {rangdari} system in the provinces of Uttar Pradesh and Bihar, in India - where although the state has collapsed an informal but reliable network of bandits have emerged to extort once and thereafter protect travellers on a specific road. In Iraq, it is perhaps increasingly difficult to know whom to bribe and whether he can offer protection against his bandit-competitor. Thus, like Washington's lobbyists, serial extortion by bandits, petty thiefs, ex-Baath officials leads to decreased amounts of efficient transactions and ultimately a breakdown. The incentives for aggrieved parties to break away from the fragile social covenant along ethnic lines and attack the sovereign (and his proxies) thus, in a sense, return to a primary 'state of nature' are rather too high.

Now think about these problems in a highly heterogenous environment; where heterogeneity emerges along relgious, ethnic and income dimensions. If violent behavior is rewarded at the margin even for a small group of people; there would be an increasing incentive for individuals to cleave along their chosen dimension. In such an environment it is perhaps inevitable that the proposal to trifurcate Iraq would rise its head in Washington and Baghdad.

Thursday, May 19, 2005

John Bates Clark Award...

On Daron Acemoglu: In Search of Lost Times...

Casual observance of modern day Economics might lead one to believe that the discipline is devoted to manufacturing apologists for the status quo or Don Quixotes who charge up theoretical windmills riding their mathematical horses. But, like all casual empiricism, beyond a notional grain of truth, this is a false conclusion. The falsity of the original observation is perhaps best exemplified by the wide variety and extraordinary rigor displayed in the works of individuals who go on to win major prizes in the discipline. Every two years, one economist is typically awarded the John Bates Clark (JBC) Award – arguably, the most prestigious (and portentous?) award in the Economics, barring the Nobel itself. This year the award was received by Daron Acemoglu of MIT for a wide range of contributions – for most strikingly his work on the role of institutions in economic growth. The JBC award is given by the American Economic Association to the best economist in the US under the age of 40. (Analogously, mathematicians/computer scientists award the Rolf Nevallina every 4 years as well, won last by Madhusudan of MIT as well.)

An obvious question is. Why is this award prestigious?

The ‘prestige’ component is a natural implication of basic sampling theory. Because of the large sample of economists (real and “policy entrepreneurs” as Paul Krugman identifies some) in the US – the distribution of quality of research is arguably continuous (but unfortunately very rarely ‘differentiable’ vis-à-vis long term implications for the discipline). That is, there are large number of remarkable theorists and empiricists on either ends of the quality spectrum; unlike say in the Economics Association of Sri Lanka or Catalonia owning to their small samples to start with. Thus, prestige emerges by virtue of being conditionally selected from a large sub-sample comprising of other-wise extraordinarily (and sometimes preternaturally) gifted peers.

There is a second component as well. Part of the prestige arrives from the implicit assent by the discipline that this researchers' agenda deserves wider recognition. Most commentators on the JBC (see Stephen Dubner on previous winner Steve Levitt in the New York Times) have made a big deal out of the possible correlation between the JBC winners going on to win Nobel Prizes. The mechanics of how this correlation works is key to understanding how the discipline itself operates. Eleven out of the thirty winners of the JBC have gone on to win the Nobel itself. But as any decent teacher of econometrics will tell you that there is a certain degree of endogeneity in this correlation which makes any such conclusion spurious in a strict sense. Although the JBC in of itself is awarded for “promising work” and as an incentive for sustained contributions; what it more importantly does is provide marginally more visibility to otherwise hyper-competitive array of research agendas. (This blog reaffirms that.) Given this new found visibility, it is entirely conceivable that a large number of existing research work are either tweaked or new ones are spawned given the previous work. Since, the Nobel Prizes are typically given for influencing the course of the discipline and spawning a whole panoply of research ideas -- the JBC ends up, often enough, causing the Nobel prize -- conditional on the Nobel winner being an American.