Wednesday, January 26, 2005

Someone tells the good Doctor to heal himself

Today's Business Standard makes a valid point.

Prime Minister Manmohan Singh’s call to Indian CEOs to emulate the Chinese model, made at a summit last week, should raise some eyebrows amongst those who have been observing the striking contrasts between the performance of the two economies over the last couple of decades.

For, any reasonable assessment would squarely pin the responsibility for the widening gap between the two, as between India’s own potential and actual achievement, on the Indian government itself.

The glaring divergence in India between the increasing availability and affordability of goods and services produced by the private sector and the deterioration in public services (including infrastructure) is one important manifestation of government failure in India.

I am not much into this "compete with China, or perish" meme; I think it is a load of bull, but I concede that it is a bull with a purpose. Businesspeople use it quite well as a tactic for, among other things, getting the Government to do what they want it to do. As long as it serves this purpose in a socially useful way (such as improving the "public services -- including infrastructure"), I have no problem with it.

Sunday, January 23, 2005

Amartya Sen

Amartya Sen's interview appeared recently in the Hindu; the interviewer is Siddharth Varadarajan. Go read it!

Tuesday, January 18, 2005

Foot-and-mouth disease at Harvard

Larry Summers, President of Harvard University, put his foot in his mouth at a conference on science and engineering careers of women and minorities. He cites the following as possible "factors" for why so few women were on math and engineering faculties at top research universities: (a) Women are not as likely as men to spend long hours (80 hours a week!) to advance their career, and (b) Boys tend to do better in math and science in upper classes at high school.

First, take a look at a rather irreverent take on this by Matt Yglesias, a Harvard alumnus. Then, read this sharp tear down of Larry Summers and his arguments by P.Z. Myers.

I just love these quotes from Myers:

  • Guess what, Summers? Boys don’t have an “innate” tendency towards science and math. Leave them alone, and they don’t grow up into natural engineers: they become animals who like to eat and screw and scratch themselves. The most important contributor to that predilection for tinkering and building and learning is education. Any possible inherited differences are miniscule compared to the power of education and cultural biases.
  • And don’t try to pretend that socialization is minimal, when the president of Harvard can stand up and seriously suggest that many people are incapable of doing great science because they have ovaries. We don’t do research with our gonads, or our skin pigments, for that matter.
  • Grrr. Apparently, congenital idiocy is not a barrier to becoming a Harvard administrator.

The Honourable President of Harvard puts up a defence [Note: This link doesn't work anymore! See the final update, below]. In the first paragraph, he simply asserts that he has been misunderstood: "My remarks have been misconstrued as suggesting that women lack the ability to succeed at the highest levels of math and science. I did not say that, nor do I believe it". In the middle of the next paragraph, he says, "In the spirit of academic inquiry, my aim at the conference was to underscore that the situation is likely the product of a variety of factors, and that further research can help us better understand their interplay". I would say it is a very weak defence.

When you hear "further research can help us", you take it to mean "we really need to understand this problem before a solution can be found". It is this that upsets a lot of people, and I think their outrage is justified.

Update: Matt Yglesias and Bitch, Ph.D. have more to say on the Summers affair.

Update (23 Jan 2005): Here are some more hard hitting posts, with more links to commentary of from either side of the political spectrum.

Update (25 Jan 2005): Sean Carroll at Preposterous Universe has a round up, together with his own long essay on this issue.

Update (27 Jan 2005): I promise that this is the last update. Summers has taken the earlier weak apology off from his site (that's why the link to it, above, does not work); it has now been replaced by this letter, where he actually sounds a bit contrite. The final link is to a balanced post by a mathematics professor who calls herself moebius stripper....

Thursday, January 13, 2005


20 February 2004; Updates: 6 April 2004, 7 April 2004, 26 April 2004; Re-written on 13 January 2005.

Take a close look at the comments section of this blog post by Brad DeLong way back in February of 2004. The main post is just a link to an article in the magazine Wired on the phenomenon of software outsourcing, without any commentary (except for the lead-off statement that it is an excellent article). Evidently, so many people (more than 100 at last count!) went to the article in Wired, came back to DeLong's blog, and decided to rant!

As many pundits, bloggers and editorialists have pointed out, outsourcing is essentially an economic issue, in which there are winners and there are losers. What we expect from the society (the Government, the Big Brother, the Party) is to choose those policies that lead to more winners than losers.

The losers in the present wave of outsourcing, clearly, are not liking it. Fair enough. However, two recurring themes in the comments section are worth pointing out.

The first is that many feel that their Government must intervene. Some have suggested making it more difficult for firms to outsource; some others have gone even further by strongly advocating an outright ban on outsourcing. From an American point of view, a ban is probably a bad deal, since it would hamper their companies, while their competitors in other countries would not face such hurdles: think Oracle vs. SAP. What would you rather want: an Oracle which survives with outsourcing, or an Oracle which is gobbled up by SAP?

The second interesting thread is that many commenters have asked for things which they would oppose (quite strongly, I would presume) if a different part of the society is affected by a similar, largely economic, phenomenon. Take the case of outsourcing of manufacturing, a wave that hit the US in the seventies and eighties and led to sharp job losses in many sectors of the American economy. Clearly, not many people (certainly not those in the software industry) cared much for the people who lost out in that wave. As one of the ommenters pointed out, the winners -- and there were so many of them -- were very happy to be consumers of high-quality and low-cost cars and other gizmos from the East Asian nations (notably, Japan).

If only the US were a welfare state (like much of Europe), the pain from outsourcing would not have been so bad and so sharp. The largely white collar software people are ruing now.

The sharp rightward turn of the US Government has almost become, since the recent elections, a near U-turn that could take that country all the way back to the early part of the twentieth century, an era that has been termed the Gilded Age by Paul Krugman, a Princeton economist and NY Times columnist. This near U-turn implies two things: (a) issues that the rich want (and they want outsourcing) will be preferred over what the poor (or the 'relatively less well off) want, and (b) welfare protections that would mitigate the hardships will be further dented, making life more difficult for the less well off.

Some updates from April 2004:

1. Stephen Cohen and Brad DeLong have "a first rough cut on what we think about outsourcing" . In particular, they say that the political debate will die soon (within a year or so), but the economic "problem", though small right at this moment, will continue to grow. A time will come when the US will have to deal with the big Outsourcing Problem; according to them, this eventuality is is still sometime away. In other words, there is time for the US to " the social safety net, the education and retraining programs...". As usual, don't forget to check out the comments section. It too contains some interesting links.

2. Brad DeLong plays class warfare! In this post , he presents a cute little economic model that shows (admittedly in a limited way) why he is right in asserting in an earlier post that "outsourcing [is] more benign if it put downward pressure on the incomes of the yuppie rich than if it put downward pressure on the incomes of the working class". As usual, don't forget to read the comments.

3. Brink Lindsey of Cato Institute has an article on outsourcing. Hat tip, once again, to Brad DeLong.

Friday, January 07, 2005

If you are into software development ...

On the other hand, even if you are not into (but generally interested in) software, you would enjoy some of the links on this page, which contains readers' nominations for the best essays on computer software published in 2004. Joel Spolsky, the author of the blog "Joel on Software", has been commissioned to put some 30 of these essays together for a book, tentatively titled "//comments".

Thanks to Joel, I found "S5", a really cool presentation software that runs on your browser.

Joel has been enormously prolific: check out his writing at this archive.

For example, in one of the articles, Joel says that software developers *really* need to know the internals of how their programs work (Hat tip to Brad Delong). Using string processing as an example, he finally concludes thus:

...I am actually physically disgusted that so many computer science programs think that Java is a good introductory language, because it's "easy" and you don't get confused with all that boring string/malloc stuff but you can learn cool OOP stuff which will make your big programs ever so modular. This is a pedagogical disaster waiting to happen. Generations of graduates are descending on us and creating Shlemiel The Painter algorithms right and left and they don't even realize it, since they fundamentally have no idea that strings are, at a very deep level, difficult, even if you can't quite see that in your perl script. If you want to teach somebody something well, you have to start at the very lowest level. It's like Karate Kid. Wax On, Wax Off. Wax On, Wax Off. Do that for three weeks. Then Knocking The Other Kid's Head off is easy.
Read the article to find out more about the algorithm of Shlemiel The Painter!

Another article is about pricing of software. Here is his concluding paragraph:

Take my advice, offered about 20 pages back: charge $0.05 for your software. Unless it does bug tracking, in which case the correct price is $30,000,000. Thank you for your time, and I apologize for leaving you even less able to price software than you were when you started reading this.
Though his conclusion sort of leaves you in a confused state, you should read this article simply because it is by Joel!

Also check out Joel's post on Google Suggest, a really cool feature from Google; don't forget to follow his link to where the real goodies exist: slashdot!

Update: Joel has more on Google Suggest here.

Employment Guarantee Act

The Employment Guarantee Act (EGA) that the present UPA government has tabled in the Parliament has been the subject of a number of comments, op-ed pieces, blog posts. Let me try to get them in one place here.

First, the links:

M.K. Venu has an op-ed in Economic Times about the bankruptcy of some of the arguments against the Employment Guarantee Act (EGA). Let me just say that I agree with pretty much everything he says.

Another op-ed in Economic Times (20 December 2004) recounts some nice arguments for EGA.

Amit Bhaduri, in an op-ed in the Hindu (27 December 2004), points out an interesting link between EGS and the right to information bill that has also been tabled in Parliament: since EGS-related work will have to be carried out by local bodies (Panchayats), the local people should have full access to what is done and to how the money is spent. Therefore, EGS itself would be rendered rather meaningless if the locals are not supported by a strong right-to-information bill. Bhaduri chides the government for not being bold and going far with both the EGS bill and the RTI bill; he feels that both are badly watered down. T.S. Srinivasan (Yale University) has responded; Prabhat Patnaik (Jawaharlal Nehru University) has written an op-ed questioning some of Srinivasan's views.

What do I think?

My initial reaction was: though it is a good scheme, can we afford it? First, let us start with some (necessarily approximate) numbers: EGA will require (about 5 crore families x 100 days x 100 Rupees per day per family) Rs. 50,000 crores (Rs. 500 billion, or about 10 billion dollars, or about 2 percent of our GDP). Admittedly, his amount is staggering; if we further consider the possibility that 5 crore might be an underestimate of the actual number of needy (really poor) families, it is truly forbidding. My initial worries about whether we can afford such a massive expenditure appeared valid.

Here is how this large amount comes down to a more reasonable figure: (a) we could club all the previous schemes for poverty alleviation, and bring them under the EGA umbrella, we probably are going to save Rs. 15,000 crores. (b) the cash component of EGA is only 60 percent of the expenditure (with foodgrains contributing the rest). Both these considerations bring the "really new" (read additional) spending down to Rs. 21,000 crores, or just under 1 percent of GDP.

There are the usual questions: will this money actually be put to good use, will our politicians play ball in a fair manner (or are they already smacking their lips at the prospect of appropriating all this money?), are there better uses for this money, etc. If we still want to rein in our fiscal deficit, where should we try to get the additional revenue from? They are all good questions, and I hope economists and other social scientists who are grappling with them will find the right (or optimal) answers.

From a humanitarian angle alone, it is clear that EGA has its merits. Its potential benefits are truly immense: so many poor families are helped, so many children will not need to go to work, so much of foodgrains (that are otherwise rotting or eaten by rats) will be put to good use, and so on. In addition, so much of rural infrastructure will get a boost (rural roads, irrigation facilities, sanitation facilities, etc). Migration from rural regions to urban centers will come down. etc, etc.

I have not seen good arguments against the scheme, only some rants about how it will not work, how our politicians cannot be trusted with it, how we cannot afford it, etc. Atanu Dey even "[goes] out on a limb and predict that if ever this EGA is implemented, it will actually increase the level of poverty and the number of poor in India". However, he hasn't (so far, at least) given any economic / political arguments to show that his dire predictions have a greater likelihood of coming true.

Even to a conservative economist, it must be clear that EGA rests not (entirely) on economics, but on politics. Given the political driving force behind it, conservatives are probably better off by conceding this one to the lefties, and put up a fight at two levels: one is a fight in which they trade EGA for something that is very dear to them (such as, (a) getting rid of subsidies for kerosene and LPG, and in one stroke, getting rid of administrative pricing of oil and oil products completely, (b) reforms of the power sector, (c) privatization of PSU's, etc.), and the other is a more serious fight to create a good system of delivering on EGA's promises with as little a wastage as possible. For example, which level of government is best suited for implementing EGA? What are the ways in which EGA programs can be made self-selecting (i.e., only those whom EGA is meant to help will actually come forward to work under EGA)?

One final thought: many of the arguments about EGA were trotted out when the (rather limited, or should I say "targeted") mid-day meal scheme that existed in Tamil Nadu in the seventies was expanded to a near-universal status by the then Chief Minister M.G. Ramachandran. Tamil Nadu is not exactly a heavenly place now, but much of the progress (particularly in educating its girl children) it has made in terms of development indicators has been attributed to that one program.

Update (23 Jan 2005): Rajiv Dawan supports a scheme that guarantees employment to the rural folk. He argues, quite convincingly too, that the current Bill dilutes this goal in many ways. He concludes the article with this:

National guarantees for the poor cost money, for which the poor agree to work on public projects at minimal wages. If designed properly, the NEG Bill will be a landmark. Otherwise, at the next election, the UPA Government will learn the consequences of making false promises to the voting poor.