Friday, January 26, 2007

Understanding the Emergency of 1975-1977

Capitalism and freedom

Great reading on Milton Friedman, by Gary Becker, Samuel Brittan, Brad DeLong, Paul Krugman, Ila Patnaik, Richard Posner, Larry Summers, Martin Wolf. For me, personally, growing up around a cancerous State in India in the 1970s and 1980s, Friedman's ideas obviously resonated. Capitalism and Freedom remains one of the most important books of my life.

Understanding the Emergency

For all of us who are used to Indian democracy and freedom, and the success of India, the Emergency (1975-1977) was a shocking failure. For those 18 months (roughly 2.5% of the life of post-independence India), we descended to the ranks of the ordinary authoritarian third world country. I think India's slide into socialism and the loss of liberty makes for a great illustration of the themes of `Capitalism and Freedom'.

Sometimes, it is felt that Nehru and Mahalanobis converted India into a socialist country. However, a more careful reading of the evidence suggests that in the early years, there was only a thin overlay of a State with a relatively free economy. There was even a sense amongst early planners - supported by views in the private sector - that the role of the State was to do the stuff that the private sector could not do, such as large capital-intensive problems which were out of reach of the then-fledgling financial sector. The process of taking away economic freedom did begin pretty early, when capital controls came in, but for a while after that conditions did not deteriorate sharply.

Indian economic policy really turned sharply towards socialism after Indira Gandhi became PM. See two articles on this by Shekhar Gupta and Ila Patnaik. In a fascinating recent article Party linkages and economic policy: An examination of Indira Gandhi's India, Charles Robert Hankla argues that the `Congress system' involving a party with a deep set of linkages to the populace was (a) breaking down and (b) controlled by `the Syndicate' and not Indira Gandhi. The license-permit raj she chose to emphasise represented a short-term political strategy to use the levers of economic policy to reward supporters and punish opponents in an environment where she was politically weak.

In the Indian case, then, we see two facts: (a) economic freedom was greatly curtailed from 1966-1975 and (b) political freedoms were curtailed in 1975. In the spirit of `Capitalism and Freedom', we wonder: Is there a link between these two facts?.

By 1975, the State was very, very powerful. The State controlled the profit rate of every firm in the country. I believe this made it difficult for people to criticise the government. There were isolated elements of rebellion in the media like Indian Express, but they were the exception. I think socialism enabled a fearful silence in the public space, which set the stage for the Emergency.

We can actually discern a two-part influence of the economic policy mistakes of the 1966-1975 period. From 1966 onwards, the economic policy got worse, so growth suffered, giving a discontented population, and raising a political challenge to the administration. But at the same time, this same bad economics enabled the administration - that controlled the commanding heights of the economy - to announce the Emergency.

Looking forward

Looking forward, the power of the State has been greatly diminished. The profit rates of a large number of firms are completely unaffected by the State. This has surely enabled greater dissent; more people are willing and able to speak out. This helps reduce the probability of a future Emergency. And, the level of discontent is lower because better economic policy has given higher GDP growth.

India now appears to be well placed to harness the synergies between expanding political freedom and expanding economic freedom. Expanding economic freedom is giving expanding political freedom, and expanding political freedom is placing restrictions on the ability of the State to take away economic freedom.

As an example, India has a very embarassing set of restrictions on the media - a role for the State that no liberal society should accept. But thanks to growing prosperity, millions of households have the Internet and are able to access all the newspapers, radio and television that the government tries to block. Citizens have greater voice so if the government tries to take away economic freedom - e.g. making the rupee inconvertible on the current account - there would be an uproar which no administration can countenance.

In China, there is a fundamental tension between growing economic freedom and growing political freedom based on the question about what will happen to the CPC. In India, my sense is that incremental progress on both fronts is feasible and synergistic; there are no fundamental roadblocks that inhibit the emergence of India as a liberal society in the 21st century, where the term `liberal' applies to both politics and economy.

Thursday, January 25, 2007

What is wrong with Bombay real estate

There are few cities in the world where the situation on real estate is as bad as Bombay. Prices are sky-high and the quality of facilities is poor. Andy Mukherjee has a fascinating article on Bloomberg about this, where he says:

Draw a circle with a radius that takes in 25 kilometers (15 miles) from the middle of Mumbai's financial district. Sea and water account for two-thirds of the surface area of this circle, compared with 22 percent for Jakarta and 5 percent for Seoul, according to Alain Bertaud, a Glen Rock, New Jersey-based urban- planning consultant for the World Bank.

All major world cities where the geography restricts the availability of land tend to make up for it by growing tall. That's where Mumbai is an exception. Planners confronted a constrained city and flattened it.

The tool they used to create a stunted city is the floor space index, or FSI, which measures how much saleable area a builder is allowed to create on a parcel of land. A higher reading means taller buildings. The index, which ranges from 5 to 15 in most Asian cities, stands at a measly 1.33 in downtown Mumbai. In the suburbs, the figure drops to 1.

A sprawling city is easier to manage. Small piles of garbage, spread across the city, can be left to rot for longer than would be possible if the trash accumulated in large, stinking heaps. More importantly, when the municipality keeps floor-area restrictions tight, it has power over developers. This power can easily translate into bribes.


Global banks, brokerages and successful Indian companies are all jostling for rentable space in dilapidated buildings, which should have been torn down and built anew a long time ago.

The reason these eyesores still exist is because they were built before 1964 when the permitted floor space index was 4.5. Today, if a developer were to buy the land to erect a new building, he would lose most of the marketable area.


The supply of land has to rise to contain housing price escalation as well as to lower office rents, which are the third- highest in the Asia-Pacific region after Hong Kong and Tokyo.

That would require a repeal of the Urban Land Ceiling Act, a law that has ended up promoting what it was supposed to prevent: hoarding of real estate.


The time for half measures is over.

A floor-area index reading of at least 5 is required across Mumbai.

Mumbai needs to rise from the ground. The question is whether civic authorities will have the will and the ability to cope with the peculiar demands of a tall city.

Wednesday, January 24, 2007

Easing capital controls on outbound FDI

In recent months, there has been a considerable focus on outbound FDI by Indian firms. Shyamala Gopinath, deputy governor of the RBI, has a speech documenting the evolution of policy on these questions.

I think there is more going on than is portrayed in this speech. The Tata Steel / Corus (would be) transaction is an example. It involves clever fund-raising in Singapore. The scale of resources involved greatly exceeds the flows across the Indian boundary as portrayed in Indian BOP data and as constricted by existing RBI rules.

When Indian firms become multinationals, capital controls bind less for them, for they will be able to transfer-price cash to a tax-efficient location where there is convertibility. Decisions from the headquarters about global FDI will then use the resources controlled at these foreign locations. The overall impact upon Indian de facto capital account openness, of "a gradual opening up" to outbound FDI is much larger than meets the eye.

This point falls within the larger theme of capital controls being porous and essentially un-enforceable in a world of modern trade, finance and MNCs.

Breaking the ice on pension reforms

There was an interesting development on the pension reforms front. A meeting of a few cabinet ministers and all state chief ministers took place, and underlined the broad-based political consensus on the New Pension System (NPS) and the PFRDA Bill. There now appears to be more clarity on moving forward with implementing the NPS without waiting for the PFRDA Bill.

What happened? See official statements from the PM, the FM, and from the meeting.

What do people make of this? Editorial commentary in Business Standard, Indian Express and Economic Times, The Times of India.Mukul Asher wrote in DNA: part 1 part 2, and Gautam Bhardwaj wrote in Business Standard. The one group that doesn't like these developments is the trade unionists. One link on the net says:

The Left parties today warned the UPA government that if it went ahead with its proposal to privatise the Pension Funds, it would be "risking" its very existence.


The top leaders of the CPI, CPI(M) and Forward Bloc made it clear that they stood for " a single pension scheme" which would ensure 50 per cent of the last pay of the salary drawn of the government employee.

Talking to UNI, CPI leaders A B Bardhan and Shamim Faizi said," The PM and the FM should realise that future of government employees and their savings can not be risked in the Stock market." " Any move to privatise the Pension Fund and handing over it to the private players will be resisted by the Left parties and the Manmohan Singh government should bother about its very existence before risking the future of the Provident Fund," they said.

The CPI leaders further made it clear that this was an" acid test for the government's relations with the Left." CPI(M) Polit bureau member and senior leader M K Pandhe said the left did not approve of a multiple Pension scheme." Paying pension to the employees, it is the responsibility of the government, which it wants to abdicate now. Where is the need for a separate agency for the purpose?" Mr Pandhe, who is also the CITU President, decried that the Prime Minister did not speak even a single word on the Left demand for a single Pension Scheme which, he said, would ensure 50 per cent of the last pay drawn of the government employee.

He said there was no justification in the contributory Pension Scheme as had been imposed on the employees since January 2004 whereby they have to contribute 10 per cent of their salary towards the funds collected in the Public Account, the CPI(M) leader said, The same should be diverted to the public Private Fund." This way the matter can be settled." CPI(M) Floor leader in the Lok Sabha Basudeb Acharia said the Left had submitted a number of suggestions to the government in this regard." But the government seems to be out to privatise the pension Funds, which we will resist in and outside the Parliament." CPI leader D Raja wanted to know who would be held responsible and accountable if the funds invested in the speculative market were lost as had been the case in several foreign countries.

But Mr Raja insisted that the discussions with the government still remained " inconclusive." Forward Bloc National Secretary G Devrajan said any social security scheme had to be "defined" and could not be decided by the market forces.

But it's a bit more complex than that. The left parties got roughly 5% of the votes in the last general elections, and trade union members probably accounted for a tenth of these. Within the left parties, Tripura and Kerala appear to have broken ranks with the trade union leaders. And, there are some signs of the Congress having the bill pass with help from the BJP [link].

For those seeking the back story leading up to these events, there is a great article by Surendra Dave. You might like to look at a blog entry on the pension reforms saga (January 2006), the backdrop of fiscal distress (September 2006), difficulties of pension guarantees (November 2006), and the problems of informal sector pensions (December 2006).

Wednesday, January 17, 2007

Difficult questions about bond market

Manish Sabharwal of ICAP has an excellent article in Economic Times today. Here is the original, but the core is more readable here:
With no disrespect to the honourable intentions of experienced people, I'd like to have an honest disagreement that the current heavy hand of policy, arguably successful in equity markets, may be inappropriate for birthing a vibrant wholesale fixed income market in India. The two mega bond market policy initiatives of recent times - the negotiated dealing system for gilts (NDS) and BSE posttrade corporate bond reporting - reflect a willingness to create monopolies, distrust global experience and raise more questions than they answer.

Let's start with NDS that has successfully introduced electronic gilt market. It is unclear whether other options for going electronic were fairly considered (ICAP's application has been pending for three years) and thus the important question is not ownership (private v/s public) but market microstructure (competition v/s monopoly).

Was NDS really important enough to overlook the opinion of the Attorney General of India that it is an exchange?

Isn't fragmenting markets by regulators unnecessary?

Shouldn't posttrade infrastructure like CCIL be open architecture and agnostic to what happens pretrade?

Why aren't regulators encouraging competition by allowing hybrid electronic intermediary platforms that would reduce transaction costs and improve price discovery?

Why is CCIL being allowed to use policy as a competitive strategy?

The second initiative is the mandatory corporate bond reporting platform operated by BSE. This is an admirable attempt to create post-trade transparency in a market whose volumes are a rounding error in our capital markets . It is also a market that is critical to funding infrastructure and lower banking and macroeconomic volatility.

But could the post-trade reporting platform provider have been neutral?

Why was BSE chosen over other exchanges that have more debt experience?

Why should BSE be given an unfair advantage in the battle for a trading system?

How will this trading system be different from the existing but famine-struck WDM platform of NSE?

Can and should banks be forced to become members of exchanges when fixed income deal execution is a small part of their overall activity?

Rather than issue fatwas that navigate the market towards a fixed destination or design, could we allow for more evolution? Philosopher Daniel Dennett called evolution a general purpose algorithm for creating "design without a designer" . Evolution chooses designs by trying out a variety of candidates: those successful are retained, replicated and reproduced, while those unsuccessful are discarded . Through repetition, evolution is a method of searching enormous, almost infinitely large spaces of possible designs for the small fraction of designs that are 'fit'.

As Dennett puts it, evolution finds "the needles of good design in haystacks of possibilities" . There is too much at stake to not allow the exploration of all possibilities (OTC, exchanges, hybrid intermediation, order matching, etc) in our fixed income market microstructure. Innovation is not the child of intelligent design (and monopolies) but of evolution (and competition).

If regulators judge policy by results, should they not be agnostic in their search for the right fit? Mumbai can and must fight Dubai's publicly-stated ambitions to "put our clock between Hong Kong and London on the walls of the world's financial institutions" . But we can't do it without vibrant bond markets and the "animal spirits" that come from brutal but fair competition.

Mission GST

P. Chidambaram has promised a Goods and Services Tax (GST) will be in operation by 2010. The idea of the GST has been kicking around since 2004, when it appeared in the Kelkar FRBM Task Force report. This requires a lot of work in implementation now, in order to achieve a rollout on 1 April 2010. A host of interesting writing on the subject has appeared lately:

  1. Amaresh Bagchi and Satya Poddar wrote a three-part article in Economic Times [link to pdf].
  2. On 7 November last year, M. Govinda Rao wrote a piece titled The path to domestic trade tax reform in Business Standard [link to BS website] [alternative].
  3. On 13 January, Sukumar Mukhopadhyay wrote an article titled Killing the goods and services tax in Business Standard [link to BS website] [alternative]
  4. Vijay Kelkar and I wrote a piece in Money Today [link to pdf].
  5. Finally, I have a piece in Business Standard today titled Mission GST.

Sunday, January 14, 2007

Risk management at exchanges: thinking beyond VaR

The risk management of the NSCC is greatly respected owing to episodes like May 2004, when volatility skyrocketed but the systems worked fine. Jayanth Varma, who did a lot of the early thinking leading up to this risk management system, did a talk at ICRIER on the 9th, as part of their finance seminar series. He spoke about new ideas in risk management, and their possible application to the task of the derivatives clearing corporation.

Friday, January 12, 2007

Enlightenment infrastructure

Will Hutton has written of the importance of `enlightenment infrastructure' in economic development, in a Chinese context:

My hypothesis when I began was that China was so different that it could carry on adapting its model, living without democracy or European enlightenment values. I have changed my mind and now see more clearly than ever the kinds of connection I identified in The State We're In between economic performance and so-called 'soft' institutions - how people are educated, how trust relations are established and how accountability is exercised (just to name a few) - are central. They are equally important to a good society and the chance for individual empowerment and self-betterment.

India made one important step forward on enlightenment infrastructure when the Supreme Court ruled (yesterday) that 288 laws, which had been placed into the "Ninth Schedule", can be challenged at the Supreme Court. Earlier, the "Ninth Schedule" permitted greater disrespect for fundamental rights on the part of elected representatives. Pratap Bhanu Mehta has a great opinion piece on this in Indian Express.

This ruling appears to reduce the opportunities for the State to infringe on personal freedom; it underlines fundamental rights as things that Parliament cannot violate. It is interesting that some of the baggage of illiberal laws that holds back economic progress in India happens to be in this set : it includes the Essential Commodities Act, the Foreign Exchange Management Act, etc. As Ashok Dhamija points out in a comment, this ruling doesn't particularly help in dealing with these.

Update: An edit in Business Standard on 26 January says:

Thanks to the Supreme Court’s judgement about its power to review laws placed in the Ninth Schedule, the Tamil Nadu government has demanded that the Constitution be re-written. The Congress, Left parties and PMK have supported the idea. When Peter Alphonse, a Congress party member of the legislative assembly, is reported to have said that “his party would not be averse to rewriting the Constitution to protect the rights of people belonging to BC, MBC, SC and ST communities (political shorthand for all the deprived communities and minorities)”, the Congress High Command did not issue a statement to the contrary. The JD(U) has also supported the demand.
One way to view this is to dismiss it as grandstanding by the allies of the Karunanidhi government, the real target being his opponents and not the Constitution. Another is to be less dismissive and ask if the fierce competition amongst fragmented political parties may lead to widespread support for the idea of “re-writing” the Constitution. If the purpose, as it seems to be, is to amend the Constitution in such a way that the Supreme Court is prevented from reviewing laws, it is clearly a terrible idea that, moreover, requires the UPA chairperson and the Prime Minister to say what they think about it. As they must know, some bad and dangerous ideas acquire political momentum because political parties take a short-term view of such issues. Two recent amendments to the Constitution—in respect of the ‘office of profit’ issue and the reservation for scheduled castes, scheduled tribes and other backward castes in non-minority, unaided private educational institutions—are cases in point.

Thursday, January 11, 2007

Barriers to migration within the country

A lot is written about inter-state disparities of income. To an economist, the twin lines of attack through which these disparities will decline are:

Connecting low price labour to markets.
This gets done by improvements in infrastructure. As an example, the new NHAI roads are linking up locations where labour is extremely cheap into global markets. New opportunities are now available for entrepreneurs to buy labour at a low price and sell goods at a global price.
Shifting low price labour to high wage locations through migration.
If the jobs won't come to the people, the people must move to the jobs. This happens through migration, which has been taking place in India on a fairly big scale (though there is some dispute on just how big it is). It is taking place in two parts - through workers that fully relocate, or through migrant workers. But in either event, these movements of people are driven by wage differentials and serve to reduce wage differentials.

These `equalising differences' are fundamental to the logic of the market economy. Both lines of attack are prominent in the Chinese growth story. India has made significant progress on the institutional transformation required for the first - this is the task of building roads, ports, airports, telecom, etc. India has not particularly begun on the second - that of becoming a migrant-friendly country. Too often, migrants lead a tenuous existence without the full rights of locals. Today, Laveesh Bhandari has an excellent article in Indian Express on the lack of institutional infrastructure to support migration.

The economic logic in favour of migration is powerful, and I expect that migration flows will grow considerably. This is partly a mere reflection of differences in the price of labour. In addition, migration is enabled by dropping costs of telecom and transportation.

There is a link between migration and the functioning of the land market. If the land market functioned better, then more land-owning agriculturists would have the choice of selling land and migrating to cities. The rigidities of the land market serve to keep poor people in place.

Update: In today's New York Times I saw an article on the problems faced by migrants in China. The situation seems to be quite a bit worse than that seen in India. Chinese migrant workers weren't able to get their children into public schools until 2003, and through some weird twists and turns, the government seems to be sometimes shutting down private schools that spring up to serve the children of migrants.

Fresh evidence on schools

Pratham did a marvellous job with their 2005 study on Indian education. They have now released the 2006 report. It is fascinating report for anyone interested in schools in India. Unfortunately, they only do rural; they do not deal with the more important issue of schools in urban areas. But their evidence is a huge step forward compared with the incentives of the government system to emphasise numbers for money spent, teachers hired, schools built and students enrolled.

Here's a comment by Sunil Jain and an editorial in Business Standard. In addition, T. Ninan wrote a related piece titled Offer an option. Update: Sunil Jain has an article in BS where he has a new table - that is not in the Pratham report - comparing private and public schools on the knowledge of children on standardised tests.

I have previously blogged about Sarva Shiksha Abhiyan and about five alternative frameworks in education.

Wednesday, January 03, 2007

Noticed an interesting paper on commodity futures trading in India

Underdeveloped Spot Markets and Futures Trading: The Soya Oil Exchange in India by Bharat Ramaswami with Jatinder Bir Singh. [pdf] They ask the question about whether the soya oil futures market in India is effective for hedging. They find that this works out well, despite all the problems of liquidity and transparency of the underlying spot market. The paper is well written and gives you insight. The econometrics is well focused - flowing directly from the questions and obtaining robust answers. It is not a mechanical exercise in crunching numbers; it is rooted in good domain knowledge about real world aspects of soya oil.

Looking back at a year of blogging

When I left the Ministry of Finance in October 2005, I resumed my earlier life as a columnist for Business Standard. I also embarked on an experiment in `the new media': writing a blog. I have made my work available through a web page on the Internet for a long time (from 1995 onwards), but blogging was new. I thought I should give it a try for a year, and then take stock about whether it's worth doing beyond. Now it's December 2006, and the blog has been in operation for a year, so it's time to review the experiment. Is it efficient for me to expend time blogging? I thought I should write some notes about my experiences with the statistics that I have, in the hope of helping others who are mulling over blogs, from the viewpoint of reading and writing them.

There are two competing visions of blogging. On one hand, the intellectual snob sees blogging as undisciplined journalism that does not pass the market test. Without worrying about wordcount, without worrying about what an editor or a referee thinks, a person is free to rant without any checks and balances. Joseph Conrad said about newspapers: Written by fools to be read by imbeciles - perhaps this applies even more for blogs. As Joseph Rago says about blogs:

Every conceivable belief is on the scene, but the collective prose, by and large, is homogeneous: A tone of careless informality prevails; posts oscillate between the uselessly brief and the uselessly logorrheic; complexity and complication are eschewed; the humor is cringe-making, with irony present only in its conspicuous absence; arguments are solipsistic; writers traffic more in pronouncement than persuasion . . .

A lot is made about the quality control that's done by google. But I am a bit skeptical about how well this works. If a lot of low quality blogs and websites link to each other, then pagerank picks up the material revered by the plebs. Pagerank was just great in the old days of the Internet, when university folk dominated the content and the links. In a situation like the academic literature, the pagerank algorithm as utilised by google scholar is great. But respecting link counts as a means to judging expertise in today's Internet? It makes me queasy.

On the other hand, blogging could be the new Gutenberg invention. The bulk of blog content is indeed petabytes of teenage angst; sophomoric tales drawing in millions of riveted sophomores. But there can be a long tail (link) of genuine intellectual content. Intellectual discussions through blogs could be a computer age implementation of the non-intermediated interaction of scientists in the 17th century.

I think there is room for optimism. As one specific example, in the field of international finance / international macroeconomics, blogs appear to be occupying a delicate and interesting middle road between newspapers and working papers. As an example, this blog entry on Christmas day is fascinating and unique. You just can't find such content if you aren't reading blogs. Conversely, until blogging came along, this middle road between the newspapers and the working papers did not exist.

Why blog?

Why write a blog? In India, we don't have the intellectual life of a Western university. I am a recluse. In my normal life, I don't meet too many people. I thought that by exposing my ideas and interests in a blog, I would hear from people who are thinking about similar issues.

In terms of sheer dissemination of ideas, it felt like a bright thing to get an exposure to the growth of the Internet in India. For more and more young people, google is the first port of call when knowledge is needed. There are perhaps 10 million people in India who read the web regularly. This number is growing at a rapid pace - perhaps doubling every five years. So even though blogging is slightly exotic today - and many people doing economics and public policy do not presently read the web - it seems like a good idea to lay the foundations for a future where everyone in India is a user of the web and quite a few people use RSS feedreaders also.

When I first created a home page on the web in 1995, that was a bit exotic, but over the years, it has become clear that this was a good bet to have placed. Starting a blog, ten years later, sounded like a similar bet to place.

I have tried to curtail the time spent on the effort, focusing on the bare core of getting some ideas out and avoiding all complexity in doing so. My apologies if the page looks bare, lacks the obligatory blogroll of the (97) RSS feeds that I read, says nothing about me, etc. Similarly I have no time for google ads. I have tried to put in the minimum possible time on the overhead, so as to minimise the opportunity cost of the blogging experiment. My main business is to think; blogging is a tool for discussion and dissemination. To paraphrase Leo Rosten, the only good reason to write a blog entry is that you can't help it.

The experience of blogging

I think the main goal - of discussions and debate with faraway people - has worked pretty well. Some of this traffic is in the comments, but a lot of it is in personal email. Somehow, many people (particularly those high up the food chain) are not comfortable writing comments on the blog. But through either channel, I have benefited in three ways. Sometimes, I get pointed to new work that I might otherwise have missed. Sometimes, I get persuaded that I was flat wrong on something. I love it when that happens. Sometimes, I get forced to respond to disagreement and clarify an argument, which is good for improving my thoughts.

As I said, this was an experiment in new media. There aren't many people in the academic or public policy world in India yet, who are blogging. Jayanth Varma is the first name that comes to mind, and I have enjoyed discussing many issues through the blog as the medium. Perhaps, over time, there will be more people blogging and we will see more debate intermediated through the blogs. The development of such an `ecosystem' would also help in terms of having more links running across the blogs, which would foster the google style meritocracy. Currently, because there are few Indian blogs, there are few links going around, and that makes it harder for a blogwriter to get noticed.

I found it's relatively inexpensive for me to write blog entries. When I write for Business Standard, there is the discipline of 1050 words: I have learned to look out for ideas that fit with 1050 words: the idea shouldn't require more words or less words. In blogging, I'm free to write 100 words or 1000 words (or more). That makes it easier to write. My entries are of two kinds. Sometimes I am merely pointing the reader to something nice that I read. At other times, I felt something about a question and I'm proposing an idea or an argument which I hope is new. In order to minimise the time expended on blogging, I try to not expend words on stuff that is already well understood and agreed-upon, amongst thinking folk. You won't find me saying that labour reforms in India are very important.

Over time, I have gained confidence and am writing more of longer and more-substantive pieces on the blog. Another simple source of bigger wordcounts, of course, is the greater inclusion of external text for the purpose of writing commentary, and making more self-contained and hence readable blog entries. I seem to have done this less in the early months.

In my experience, blogging fits the old-style intellectual better, as opposed to the modern specialise-and-publish mode of academics. A good academic today is supposed to focus on one thing, and specialise and specialise to a point where he knows everything about a narrow field. Speaking for me, I have always had a lack-of-specialisation syndrome. I like to survey a field from high above, absorb the big picture, and make links across fields. Blogging has accentuated my lack of specialisation; it has increased my breadth and that comes at the price of depth. If your ambition is to be a successful academic, this is probably not a bright thing to do.

Blogging has helped me by giving me `an intellectual diary' of my life. I tend to write about the stuff that I'm reading and the places where an idea strikes me. In the world before blogging, these experiences would rush through my brain and then subside, and retrival of old information and old ideas from the human brain is hard. I'm getting old, and my in-core searches aren't as good as they used to be. The blog serves as a searchable log of what I was thinking and what I was writing. Of course, blogging helps in taking a rough idea into something written down, which always improves clarity. I remember once, I thought I had a great idea on how to do municipal wifi, but when I started actually writing it down, I realised the idea was broken.

Later on, when something comes up, I find myself saying "I faintly recall reading something by XXX about this; I had blogged it; let me look it up from my blog". This often happens when I am writing email - instead of writing a few paragraphs of an argument, I find I'm often able to just google it in my own blog and serve up the URL. So blogging has helped me to improve my information management.

Some statistics about one year of blogging

Usage of the blog: I only setup measurement through sitemeter from July 2006 onwards. The data shows:

MonthVisitsPage views
July 1,3002,200
August 1,6002,800
September 2,0003,600
October 2,0003,500
November 2,6004,500
December 3,0005,200

I find it useful to think that in 24 hours of a day, if there are 10 pageviews an hour, this is 240 pageviews a day or 7,200 pageviews a month. So these statistics reflect a utilisation of under 10 pageviews per hour, as yet. These numbers are small when compared with the successful economics blogs of the West, where 240 pageviews a day isn't much.

These numbers for usage surely overstate the genuine utilisation of the blog, for every now and then, we landup at some page through google, and at a glance, we know it's not right. But those users get counted as `visits' or `page views'. When pages views exceeds visits, it suggests a person who wandered around, so it's less likely that it was a wrong number.

A certain growth of the usage of the blog is inevitable and non-informative. On one hand, the stock of content on the blog goes up linearly through time, which leads to more search engine users landing up on some blog entry. In addition, there is a geometric growth of Internet usage in India. These two factors should give strong growth of the usage of the blog, regardless of how much sense I am writing.

Where are the readers?: In the early days of my personal home page, there was a skew in favour of readers outside the country, since India had problems in telecom policy. Now, it looks like the majority of the readers of my blog are in India. The data shows: 51% India, 27% US, 4% UK, and then other smaller shares.

RSS subscribers: I first used an RSS feed that is produced by blogspot. But they don't give me interesting information about utilisation. That feed URL continues to work, but in January 2006, I switched to exhibiting an RSS feed from I have data for how users of this (feedburner) feed has grown. It seems to have added an average of roughly 10 subscribers per month, and has roughly 120 subscribers at the end of a year.

Users of the RSS feed make up only a small fraction of the usage of the blog. The bulk of blog users seem to either landup by searching for a string on google, or going to google and looking for "ajay shah blog".

Users requested me to modify my blog settings so that the full blog entry appears in the RSS feed. It is possible for a user to then only use an RSS feedreader and never show up at the blog.

Choice of service provider

Anyone who thinks of writing a blog has to think of a place where this will be implemented. Overall, I think blogspot has been a good provider:

  • I do my authoring by writing HTML files and emailing them to a stated address using a very convenient one-liner:
        $ mutt -a file.html -s "Title" address < /dev/null 
  • Blogspot has had severe performance problems. Hopefully they're google and know how to solve them.
  • The insertion of .jpg files into a blog entry is painful.
  • Blogspot does not give out any statistics about usage, RSS feed subscriptions, etc.
  • Finally, the blogspot search is extremely broken, to a point where I have experimented with offering alternative search facilities on my blog. Hmm, they're google - shouldn't they know how to search!?

So I would say that while blogspot is good, it's not yet there.

I feel relatively safe dealing with google on various kinds of evil things like my blog vanishing suddenly when a government gets unhappy. One of the nice things about blogspot is a sensible, almost-hand-written structure of URLs, which makes it easy to do a recursive wget. Just to be safe, I periodically do a wget through which the entire blog is backed up on my notebook. And, because the blogspot search is very broken, I hunt down stuff on my own blog by grepping within these files.

If you find this look back at blogging to be of interest, you might like to read an article I wrote in Business Standard, looking back at the newspaper opinion piece, when I moved into the Ministry of Finance in 2001. It ran under the title Last ride together. In the event, this proved to not be the last ride :-) but only a four-year hiatus.

Tuesday, January 02, 2007

Beware the tranquility

Global equity volatility and Indian equity volatility is low. The blue line is the implied volatility off the S&P 500 options market; the red line is the backward looking rolling window volatility of Nifty (width = 66 days).

I wrote an article in Business Standard titled Beware the tranquility. In this, I worry that the world could undergo a `phase transition' and jump back to a configuration of high volatility, high risk premia and high liquidity premia. This is linked to the arguments made by Raghuram Rajan in June 2006 where he draws a link between low interest rates and the remarkable decline in risk premia and liquidity premia.

Larry Summers wrote a somewhat similar cautionary piece on 26 December titled Lack of fear gives cause for concern. Kenneth Rogoff discusses possibilities for the equity premium on Project Syndicate. And if you like to see these things on the sweep of history, look back at the First World War.

Monday, January 01, 2007

Faulty measurement of the economy

In thinking about India, a big constraint lies in the statistical system. Many key pieces of information are mis-measured, and appropriate skepticism needs to be applied when consuming official statistics. As an example, in my experience with learning macroeconomics in an Indian setting, it took many years to gain comfort about what time-series one is going to take seriously. Mundane issues like measurement of inflation, which ought to be effortless in a mature market economy, are a complex challenge in India.

There are three datasets where I feel there is a big gap between perception and reality: the National Sample Survey (NSS), the Annual Survey of Industries (ASI) and "State Domestic Product" (SDP). I find myself being extremely skeptical about these three. Today's Business Standard has a good article by Sunil Jain about the measurement problems of NSS (and NAS).

Many academic research papers merrily proceed on trying to use these datasets. When I read these papers, I generally end up shaking my head saying "I just don't trust any of this". It's a shame, for these datasets really ought to be done better. But it's also a waste when researchers indiscriminately proceed to expend man-months of time on weak foundations. I think anyone working on India should be a little more thoughtful about difficulties of measurement. To a discriminating reader, weak data foundations lead to a fatal loss of persuasion, and if a research paper won't persuade, why bother writing it?