Friday, May 27, 2011

Recruiting the right MD for the IMF

The recruitment of the IMF MD has turned into quite a controversy. For an interesting set of views, see this page on the website of The Economist. In a remarkable development, the EDs of India, China, Russia, Brazil and South Africa came out with a clear joint statement on the silliness that is afoot.

There are four perspectives on this question which are worth noting:
  1. There is an obvious gap between the power structure at the IMF, which reflects the structure of the world economy after the Second World War, as compared with the present reality. As an example, at present, the Netherlands has 2.08% of the votes while India has 2.35%. But Indian GDP is now $1.6 trillion while Netherlands is at half that.
  2. The world would benefit from a competent and capable IMF. The best man (or woman) for the job will not be obtained by having any restrictions on nationality. As an example, in today's world, a name that leaps out to me is Stan Fischer. But he's not European, and hence was never even considered for the top job in the last decade. (As with Montek, he is now over age 65 and is hence not eligible for the job today). Given that a large fraction of the top economists of the world are not European, this rule yields a less capable IMF.
  3. A quota system where the IMF MD must now be from an emerging market is as bad as a quota system where the IMF MD is only recruited from a European country. The key is to get away from all these restrictions, and to only recruit the best person for the job. The emphasis should be on technical capability. As an example, see how in the UK, they recruited an American into their Monetary Policy Committee.
  4. In the standard narrative, one hears the idea that in this crisis in Europe, the Europeans are gaining from their control of the IMF. I disagree. In the Asian crisis, it was good for Asia that the IMF was not conflicted by considerations of domestic Asian politics. Similarly, the IMF program in India in 1981 and 1991 was uncontaminated by domestic Indian political considerations. This helped produce technically sound programs, which helped in jumpstarting India's growth. It is not accidental that we see structural breaks in India's GDP growth around these two dates.
    What Europe needs most is a tough IMF, which will be a stern taskmaster, which will force difficult political choices so as to heal the economy. Economic policy in Europe today needs to be cruel to be kind. Instead, by placing a string of career politicians from France into the IMF MD's job, the valuable role which the IMF could have played in solving the European Crisis is being negated. This damages Europe. The wise thing for Europe today is to say: Give us a tough and competent taskmaster, recruited through global search, where European politicans are not allowed to apply. The biggest loser from the present arrangement is Europe.

Monday, May 23, 2011

Opportunities in the Macro/Finance Group at NIPFP

The Macro/Finance Group at NIPFP has opportunities in several roles.


We require individuals with a Ph.D. in economics or finance, with an interest in original research in our fields. These would be contractual appointments for a period of one or two years. One or more publications in international journals would be helpful, as would be the ability to carry research from inception all the way to publication.

Policy research associates

We require people who can participate in large complex research projects in the field of public policy. As you will see at the URL above, the Macro/Finance Group at NIPFP offers many opportunities for a meaningful engagement with government in the ground realities of India's economic reform.

Policy work is highly inter-disciplinary. The policy group at NIPFP draws together people with a knowledge of public economics, public administration, financial regulation and law. We welcome interest in these positions by people with strong capabilities in one or two of these areas, and curiosity about the others. The ideal candidates would have read the Percy Mistry and Raghuram Rajan reports, and have familiarity with the things being talked about in this blog.

Quantitative research associates

We require individuals with a Masters degree in economics, finance, public policy or statistics. The work involves participating in academic research projects in the fields of macroeconomics and finance, and practical macroeconomic policy analysis. Desirable features include: domain knowledge; knowledge of computer programming, ideally in R; experience with CMIE databases and datastream.

Research programmers

We require a senior person who would play a dual role. On one hand, he would be responsible for an existing system which includes linux desktops, linux servers, and a project management system based on redmine and svn. This is expected to take up roughly 20% of effort. The prime focus will be participating in development work of complex analysis of economic and financial data. This development work is primarily in R. It includes building internal tools and also some packages released as open source.

We also require a junior person who will primarily work on systems administration for an existing system which includes linux desktops, linux servers and a project management system which uses redmine and svn. Some Microsoft windows knowledge is also required. On a second priority, there would be development in R.

The Macro/Finance group at NIPFP is a research environment: non-hierarchical, low political complexity, high IQ people, high intensity and involvement. Such work could be particularly appropriate for a person who is at present a computer engineer but desires knowledge of economics and finance.


We require an experienced program manager who would be able to handle budgets, financial reporting, deadlines and deliverables, and client interactions. Experience with government procedure would be a positive.

Generic features

The Macro/Finance Group at NIPFP is a conducive research environment including a modern office. Compensation is generally superior to that seen in government academic institutions. The policy and the quantitative teams are strongly interconnected with significant spillovers of knowledge.

If any of these interest you, please contact Anurodh Sharma (anurodh54 at with your resume by 31 May 2011, where you clearly identify which of these roles is of interest to you.

Friday, May 20, 2011

Interesting readings

What is right about the CPI(M), by Prabhat Patnaik, in the Telegraph (16 September 2010).

Thomas E. Ricks (in Foreign Policy) and Lawrence Wright (in New Yorker) on Pakistan.

C. Rangarajan on the debate about the debt management office and about inflation targeting (the latter is an interview with Tamal Bandyopadhyay).

Saurabh Mishra, Susanna Lundstrom and Rahul Anand have a fascinating piece on the sophistication of India's service exports. Many people suffer from what I call `the widget illusion', where somehow it is good to make tangible things, and making intangible things is considered wrong. It is high time we get away from such notions.

Kenya's experience with mobile phones and payments is important for us in India. Read William Jack and Tavneet Suri on this, on voxEU.

I found there are interesting links between this article in The Economist, and the ideas on system-driven credit in a UID world in this committee report.

Do you use up the power of monetary policy to stabilise inflation, or do you use up this power to manipulate the exchange rate? Some people think that manipulating exchange rates, and thus fueling export growth, is a shortcut to high GDP growth. Nicolas Magud and Sebastian Sosa (on voxEU) say that the potential payoff from exchange rate misalignment is small.

A working paper: Liquidity considerations in estimating implied volatility by Rohini Grover and Susan Thomas.

A working paper: Improving the legal process in enforcement at SEBI by Dharmishta Raval.

A working paper: Has India emerged? Business cycle facts from a transitioning economy by Chetan Ghate, Radhika Pandey, and Ila Patnaik.

Mobile trucks that sell food, and link up to customers using twitter: is India is ready for this? See Caroline McCarthy on CNet News.

A first response on the killing of UBL by Steve Coll.

Robert S. Boynton has an article in the Atlantic about how modern communication technology is actually making a small difference to breaking down the North Korean government.

Henry Shukman has a great story in Outside magazine about the 3000 square kilometres of `Chernobyl Exclusion Zone' which has turned into a miracle for biodiversity. I often wonder what would happen if 3000 square kilometres of prime Gangetic land became true forest.

Perhaps 10% of blind men can teach themselves how to see.

Michael Lewis has a persuasive sounding article, about how a Richter 7.9 earthquake that hits Tokyo will devastate the world economy. This was written in 1989. By and large, these things did not happen in the recent Richter 9.0 earthquake. Yes, the recent quake did not frontally hit Tokyo, but then Richter 9.0 is way bigger than 7.9 (this is log scale). It is a useful exercise, for everyone interested in finance, to read this article and understand how such journalistic thinking goes wrong.

Is research funding going into randomised trials yielding a good bang for the buck? My personal view is that a better use of money is to build datasets like this which are then placed into the public domain, and used by hundreds of researchers.

Thursday, May 19, 2011

Slamming the accelerator while hitting the brakes

The #1 problem of Indian macroeconomic policy is the inflation crisis. From February 2006 onwards, in every single month, inflation has exceeded the target zone of four to five per cent. I'm measuring inflation as the year-on-year change of the CPI-IW. The latter is the best measure of the overall price level in India today.

This macroeconomic instability is damaging the ability of economic agents to plan and invest for the future, because it's hard to envision interest rates and prices when faced with such high uncertainty. High inflation thus damages growth.

Many people in India like to make excuses about inflation. One day, inflation is about the price of onions. Another day, inflation is about a global commodity shock. Many people like to open up the sub-components of WPI and explain away inflation by saying "but it's only concentrated in a few things which make up x% of the overall basket". And so on. While each of these idiosyncratic factors can generate relative price changes, they cannot explain sustained price rise of the overall household consumption basket.

Sustained and persistent inflation is not an act of god. It is made by mistakes in macroeconomic policy. It can and should be contained by solving these problems of macroeconomic policy.

On May 3, Dr. Subbarao announced a fairly good policy statement. It continued to talk about WPI while the best inflation measure is the CPI. But for the rest, it was the first time that RBI was starting to take the inflation crisis seriously. And that was good. Also see an Indian Express column by Ila Patnaik on May 6.

Sadly, RBI's commitment to the problem of inflation lasted for six days. On 9 May, Dr. Subbarao did a speech in Switzerland which essentially robbed RBI's stance of credibility. Ila Patnaik has a column in the Indian Express about the damage that this speech has caused. You might like to also see this old column of mine on the problems of RBI.

Consider the date on which the rate hikes began. Compare two alternative worlds:
  • In one world, RBI says: "We care about inflation, we will do what it takes to get y-o-y changes of the CPI-IW back to the target zone of four to five per cent". And the rate is hiked by 25 bps. And this is repeated a short while thereafter. And so on. In this world, the expectations of economic agents get modified alongside the rate hikes.
  • In another world, every time RBI raises rates, RBI says "actually we are not so serious about inflation". In this world, the expectations of economic agents do not get modified alongside the rate hikes.
Monetary policy works by directly crimping aggregate demand (e.g. driving up the EMIs that people pay, or the cost paid by firms for working capital) and by reshaping expectations and thus the decisions about wage / price hikes. By damaging the latter, RBI has imposed more of the heavy lifting upon the former.

What does it take for RBI to persuade us that they are serious about inflation? Commitment to the floating exchange rate (thus removing this conflict of interest that can damage monetary policy), movement on the DMO (thus removing another conflict of interest that can damage monetary policy), and sound monetary economics going into speechwriting (and future monetary policy formulation). By failing on all three scores, RBI is generating a situation where there is no commitment that in the future, RBI will fight inflation.

Whether RBI wants it or not, India will fight this inflation crisis, which is the #1 cloud on the horizon of India's macroeconomic policy. The politicians require CPI-IW inflation to be back to the four-to-five per cent zone by late 2013, well in time for the elections in 2014. The pressure is simply going to ratchet up. The only question is about how monetary policy will fight inflation. If the instrument of monetary policy is refashioned to fight inflation, then the amount of pain that has to be inflicted through rate hikes, that is required to get the job done, will be lower. If the instrument of monetary policy is mis-managed, then a bigger set of rate hikes are required to get the same thing done.

In the medium term, RBI needs to build a team of top quality economists, who gain street cred by exuding knowledge of monetary economics. In the short term, the least that is required to be done is to stop the flow of low quality speeches.

Wednesday, May 18, 2011

Books that should be read before starting a Ph.D. in economics

Suppose a young person is going to start a Ph.D. in economics. What essential readings would you recommend prior to this?

In my opinion, the Ph.D. in economics involves a heavy emphasis on tools. But the story isn't told, about why we are building these tools. The intuition isn't built, about the world out there that we seek to model. I always joke that economics students who are clueless about reality are like a child studying projectile motion without having ever thrown something into the air.

So I thought it's useful to pick a set of books that touch on the great themes of the world, often going into troublesome terrain that the models aren't very good at, so as to lay a foundation of background knowledge and historical knowledge which can pave the way to usefully assimilating what's taught in the economics Ph.D. Of course, they should be books that are fun to read and un-putdownable.

Here's my suggested compact checklist of books worth reading. Please do suggest books, and disagree with this list, in the comments to this post.
  1. The evolution of cooperation, by Robert M. Axelrod
  2. Good capitalism, bad capitalism, and the economics of growth and prosperity by William J. Baumol, Robert E. Litan and Carl J. Schramm
  3. A splendid exchange: How trade shaped the world by William J. Bernstein
  4. The elusive quest for growth by William Russell Easterly
  5. Invisible engines: How software platforms drive innovation and transform industries by David S. Evans, Andrei Hagiu and Richard Schmalensee
  6. The ascent of money by Niall Ferguson
  7. Economic gangsters: Corruption, violence and the poverty of nations by Raymond Fisman and Edward Miguel
  8. Capitalism and freedom by Milton Friedman
  9. The great crash of 1929 by John Kenneth Galbraith
  10. The age of uncertainty by John Kenneth Galbraith
  11. Exit, voice, loyalty by Albert O. Hirschman
  12. Development, geography and economic theory by Paul Krugman
  13. More money than God: Hedge funds and the making of a new elite by Sebastian Mallaby
  14. Reinventing the bazaar: A natural history of markets by John McMillan
  15. Readings in applied microeconomics: The power of the market edited by Craig Newmark
  16. From the corn laws to free trade: Interests, ideas and institutions in historical perspective by Cheryl Schonhardt-Bailey
  17. Seeing like a State by James C. Scott
  18. The company of strangers by Paul Seabright
  19. Information rules: A strategic guide to the network economy by Carl Shapiro and Hal R. Varian

Monday, May 16, 2011

A new low for Indian economic policy

Strange things in the appointments process:

Thursday, May 05, 2011

Solving the problem of black money in real estate

by Shubho Roy and Pratik Datta.

Manmohan Singh as finance minister killed off smuggling, by eliminating customs duties. Black money in the real estate sector recently attracted his attention. He suggested lowering of stamp duties to check the flow of black money in this sector. But will this solve the problem of black money? And how will the State compensate for the loss of revenue collected from stamp duty?

Stamp duty is a transaction tax; it is charged as a percentage of the transaction value of the property. Public economics teaches us that all transaction taxes are bad taxes, that the right level for the stamp duty is zero (as it is for all taxation of transactions).

The stamp duty distorts the behaviour of parties to the transaction. Stamp duty on property is usually paid by the buyer. Hence, the buyer tries to coax the seller to agree to undervalue the property on paper (the transaction value declared to the government) and accept the rest in black money. On the other hand, sellers have an incentive in accepting black money from the buyer in order to evade the taxation of capital gains. As long as real estate capital gains are taxed, eliminating the stamp duty will influence the behaviour of the buyer but not that of the seller.

Hence, modest changes in the stamp duty rate will not solve the problem of black money in real estate. When stamp duty is eliminated, the buyer will be comfortable with zero evasion, but the seller will still urge him to take some black money.

Bad taxes should be eliminated because they are bad taxes. There should be no attempt at finding a direct replacement. As an example, India largely phased out customs duties, because the economists said these are bad taxes, without specifically trying to find a replacement. The elimination of customs duties enabled high GDP growth, and the main pillars of taxation (income tax and the VAT) generated bountiful revenues. A similar story will hold for stamp duty.

The economists teach us that all taxation of transactions is wrong. The property tax suffers from no such problems. Much work is needed in India in building a sound property tax system. In some countries, property tax revenues can be as large as 1% of GDP, which is a very big number compared with the financing of local government in India. The key issue is that the average value of property in each micro neighbourhood (e.g. a 200 metre stretch of road) needs to be assessed correctly and revised every year. This should then be used as a preumptive property tax rate, per square foot, for that micro neighbourhood. Once this is done, property tax collections will be a powerful source of revenue for local government.

There is a link between these two issues. As long as there is a stamp duty and high taxation of real estate capital gains, the reported data will understate property values. This will hamper the revenues obtained through the property tax. To the extent that we solve the twin problems of stamp duty and capital gains taxation, the data in the hands of government about real estate prices will improve, and this will bring property tax to life as a significant revenue source.

Author: Pratik Datta

Sunday, May 01, 2011

Succession problems in Indian firms

Democracies are more stable

Autocratic countries often appear to have a clean and stable political system. A government is clearly in charge. Businessmen like to deal with such a government, because you can go into a room with a powerful person and walk out holding a deal. You can do business with them.

Democracies, in contrast, are messy. The essence of a democracy is the dispersion of power. When power is dispersed between many individuals and institutions, decision making is slow and messy.

Differences are visible in public. A businessman finds it difficult to deal with such a government: He can't walk into a room and do a deal. Instead, deals (such as an airport contract or a mining concession) go through a contentious procurement process in the public domain. In the third world, the procurement and regulatory procesess are often riven with corruption, which makes a benevolent dictator look good.

While an autocracy may appear to be calm and stable, it actually suffers from two dimensions of instability. The regime suffers from the silent reproach of a million tear-stained eyes: You never know when an upheaval will come about. And autocracies suffer from problems of succession. When the strongman dies or gets killed, you never know what's going to happen next. When power and decision making is centralised, succession becomes difficult.

When the caudillo comes towards the end of his life, this triggers off instability because people around him are solving dynamic programming problems. I suspect this was part of the story of how Mubarak's world fell apart.

A business with a strong CEO is like an autocracy

Many firms have centralisation of thinking, power and decision making in the hands of one person or one family. This often looks nice for a while. There is clarity about who is in charge; the CEO is generally well incentivised; the CEO generally works hard and many such firms are highly successful.

But such firms face difficulties of succession. Precisely because so much power and decision making was concentrated in one person, it is difficult to replace him or her.

As with good countries, good companies evolve from concentrated power to dispersion of power. A good country is one in which power is highly dispersed, where thinking and problem solving is taking place in millions of places by empowered individuals who are not waiting for instructions. In similar fashion, a good company is one in which the CEO does not dominate the landscape: the board of directors (above) and the management team (below) play a much stronger role when compared with conditions of dictatorship. As with a country, if one person is doing all the thinking, the firm is capable of little. In a good firm, the energy and imagination of dozens or hundreds or thousands of people is harnessed.

For small problems, one thinker is often adequate. As an example, to run a coffee shop, one mom and pop suffices. But to run a large, complex, modern knowledge-intensive firm, we need to harness the energy and imagination of hundreds or thousands of people. When such a firm is limited to the capabilities of one person, no matter how good he or she is, that yields stagnation.

In an autocratic company, there are serious problems of succession. A dominant CEO is hard to replace even if one were recruiting from the open market. Matters are often made worse by limiting CEO search to a family. In contrast, when power is dispersed, succession is inherently safer. Even if there is a lot of sound and fury in succession, there is less that can go wrong.

It takes a long time for a country to learn how to live within the complex checks and balances of democracy. In similar fashion, it is not easy to be a sophisticated modern firm, where the CEO is not a demigod. It is a difficult transition to make, to go from an autocratic environment to a democratic environment.

I believe that political analysts, globally, make the mistake of overstating the stability of a dictatorship and underestimating the stability of a democracy. In similar fashion, I feel that many people underestimate the succession problem of a family business and overstate it in a professionally managed company.

On the other hand, agency problems

This case against family run companies is very strong, for large organisations where it is essential to have many people thinking. However, the key problem that the professionally managed company faces is that of agency conflicts. With a family company, the incentives of the CEO are clear. With a professionally run company, it is not easy to ensure that the management team works for the interests of the shareholders.

On one hand, power needs to be dispersed because otherwise we can't have hundreds of people who are empowered and thinking. But when power is dispersed in such fashion, there is the heightened danger of theft.

The management of a professionally run company is therefore all about the tension between the efficiencies (economies of scale + large number of people who are thinking) on one hand versus theft on the other. Once again, it isn't so different from the agency problems that democracy is riven with.


When a dictator is succeeded by his son, it looks like a smooth and easy transition, but it is actually a situation that is fraught with risk.

Succession at Infosys has been contentious and in the public domain. As with an Indian general election, it looks messy. But the problems here are overstated. Infosys is doing something relatively new in India: they are a professionally-managed dispersed shareholding company with disperson of power. While such succession looks messy, there is greater stability under the hood.

Governance problems of Indian firms

India is remarkable in having high quality firms. But at present, very few firms have the checks and balances of dispersed shareholding, a genuinely powerful board of directors, a professional management team, and the absence of dominant founders or family. There are a few such examples -- L&T, Infosys, ITC, Axis Bank, ICICI -- but as of yet, it is rare.

Many of the successful giant firms of the present Indian landscape are a bit like China: They look great today but they run the risk of a USSR event as they face the transitions of the future. The Indian corporate sector has a lot of work in store, in refashioning the giants of today, using the governance DNA of firms like L&T, Infosys, ITC, Axis Bank and ICICI. Those transitions will not be easy. As Lant Pritchett says: I recently did a study examining the growth consequences of sudden large democratisation (a shift in the POLITY index of more than 6 points). Of the 22 cases that experienced rapid democratisation with above average growth: (a) all but one had a growth deceleration, (b) the average deceleration was 3.5 ppa, and (c) the predicted deceleration was increasing with growth—roughly, post-democratisation countries reverted to world average growth. Transitions out of dictatorship are not easy.