But clouds got in my way: Bias and bias correction of nighttime lights data in the presence of clouds

by Ayush Patnaik, Ajay Shah, Anshul Tayal, Susan Thomas.

Night lights is an opportunity to measure prosperity, using an eye in the sky, without requiring institutional capacity in economic measurement on the ground. The first wave of research used the DMSP-OLS dataset, which had annual images from 1992 to 2013. An improvement in this field was the launch of Suomi-NPP in 2012 where the pixels are smaller (0.5km x 0.5km), and the frequency shifted from annual to monthly. A substantial economics literature has found innovative applications of this data. When research projects are set in India, most researchers have relied on the district-level dataset that is generously released by the World Bank.

In a new paper, 

• We suggest there is a downward bias in the radiance, that is associated with the presence of clouds. The magnitudes are economically significant, e.g. -28% in July for Bombay.
  
• We propose a bias correction scheme that partly corrects for this bias.
• We have released the source code which implements our improved methods and conventional methods, so they can be used in data construction by applied economists and for methodological research in remote sensing.
  

The problem of bias

As an example, consider the radiance seen at the satellite from the city of Bombay:

The red line is the aggregate radiance from Bombay. It shows peculiar annual dips. The vertical dashed lines mark July months, where the monsoon is strongest (on average). The lower graph is the number of cloud-free pixel-days that make up this aggregate radiance. There is a pattern: odd dips in radiance that are correlated with low values for the number of cloud-free pixel-days.

Is this just the seasonality of income, which happens to be correlated with the seasonality of cloud cover? 

The graph above juxtaposes the seasonal factors of monthly aggregate income in Bombay (the black line) vs. the seasonal factors of monthly aggregate radiance for Bombay (the red line). There is no seasonal dip of income in July as is the case with nighttime lights.

This is just an example, for the city of Bombay. The paper has large scale evidence about the presence of this problem more generally.

We conjecture that for a pixel, in a month with a low number of cloud-free images, even on those few days, there are light clouds which attenuate the signal, thus inducing a downward bias in the observed radiance.

A partial bias-correction scheme

When a pixel has both bright and cloudy months in the data, we are able to estimate the bias and correct for it.

There are pixels which are cloudy all through the year. Here, the bias is unidentified.

Our bias-correction scheme works cautiously, only modifying the data when there is high confidence that there is bias and we are able to estimate the magnitude of the bias. It reduces the bias but does not eliminate it.

As an example, consider Bombay:

As before, the black line has the seasonal factors of aggregate income in Bombay. The red line has the seasonal factors of conventionally cleaned night lights data. The dashed purple line has seasonal factors for the night lights data released by the World Bank. 

The blue dotted line is the new bias-corrected night lights data. These seasonal factors are closer to the black line and an improvement upon the two conventional datasets.

Once again, Bombay is just an example; the paper has large scale evidence which demonstrates these gains. For the aggregate radiance of India:

Here also, the dotted blue line (the seasonality of the new night lights data) is closer to the black line (the seasonality of aggregate income in India), and fares better than the two conventional datasets (the World Bank's release or conventionally cleaned nighttime radiance).

Reproducible research

We have released the data and R code to reproduce all our calculations for Bombay. And, we have released a Julia package using which the new tools can be used for methodological research and applications. This software consumes a pixel-level NASA/NOAA VIIRS dataset and returns a bias-corrected pixel-level dataset which will readily fit into analyses of the existing NASA/NOAA VIIRS data. This is also the first open source package for conventional cleaning.

Distribution of self-reported health in India: The role of income and geography

by Ila Patnaik, Renuka Sane, Ajay Shah and S. V. Subramaniam.

In health research, we study the causes and consequences of health at the individual level. This requires measurement of the health status of individuals. One simple path lies in asking a person: "Are you feeling well today?". This `self-reported health' (SRH) is a measure that is easy to implement, and has limitations in that psychological factors are present. A significant global literature has emerged, which draws on this measure to explore the causes and consequences of health.

The CMIE CPHS is an important new dataset which has longitudinal data for about 170,000 households, measured three times a year. They measure SRH for each individual in each wave. This measurement of SRH, alongside a rich array of household characteristics, makes possible many interesting research projects. In a new paper, Distribution of self-reported health in India: The role of income and geography, we discern some new facts and phenomena about health in India, through this data.

We use data for calendar 2018 and 2019, which works out to 3.5 million observation of a person in a wave. These years were chosen in order to obtain a baseline description of health in India, while avoiding the pandemic of 2020 and the possible impact of demonetisation in 2017.

What do we find? On average, ill health is observed in 3.25% of the records. On average, people in India are unwell for about 12 days a year. There is a U-shaped curve in age, with higher ill health rates for the young and the old.

We get a nice map of the variation of the ill-health rate across the country. This is interesting, in and of itself, as it shows us something about health care requirements. However, some of this variation reflects geographical heterogeneity in income and age structure.

We estimate logit models which explore correlations between standard socio-economic measures and the ill-health rate. The important sources of variation turn out to be age, income and location.

We then focus on an approximately modal person. Model-based predictions for the ill-health probability are constructed for this individual. This yields a map of the predicted ill-health rate --  

 

This shows the variation of ill-health in the 102 `homogeneous regions' (HRs), after controlling for income, age structure and other standard socioeconomic characteristics. It is an interesting and new map. These results do not conform with the standard stereotypes of north vs. south. Epidemiological research is required in understanding what is at work in each of the difficult HRs. Major gains in the health of the people could potentially be obtained by focusing on these hot spots and finding the right public health interventions.

We then ask: are rich people healthier than poor people? As the rich fare better on nutrition, housing quality, knowledge and access to health care, we expect there would be such a correlation. This is indeed the case in the overall aggregate data. However, there is strong geographical variation in this correlation. Ill health and poverty are positively correlated in only half of the country. There are even HRs where the relationship is reverse -- where poor people report better health than the rich. Further, the two maps (the map of ill health of the modal person, and the map of the places where ill health is not positively correlated with income) show different patterns. They are distinct phenomena that invite further exploration.

The geography of firms and firm formation in India

by Surbhi Bhatia, Manish Kumar Singh, Susan Thomas.

We look at the geographical location of firms in the figure below. The map on the left shows the stock of the firms in India, with all firms as on 31st Dec 2015 while the map on the right shows the flow of new firms that were added in the year ended 31st Dec 2016. All values are expressed per unit population.

The graphs are built using the list of all active companies registered with the Registrar of Companies on 31st Dec 2015 and on 31st Dec 2016. This is from the Ministry of Corporate Affairs (MCA) website. This data contains the unique identification number of the firm, firm name, current status (active or not), type of firm (public, private, LLP, etc.), authorised and paid-up capital, date of registration, office address, sector, and ownership details.

Most Indian firm research uses the CMIE firm database, which has deep information about 50,000 firms. The MCA dataset shows limited information about all limited liability firms. The headcount of the stock of firms on 31st Dec 2015 was 1,022,174 and the new firms registered in 2016 were 111,274 in number.

What do these maps tell us? Let us start at the left map. This is on somewhat expected lines. The highest values are found in Maharashtra and in West Bengal (where there are many ancient firms). For the rest, there is strength in the West: from Punjab/Haryana to Rajasthan to Gujarat to Maharashtra to the Southern peninsula. This is generally the region of prosperity in India, in the common preconception.

The map on the right, with new firm creation in 2016, diverges from this picture in striking ways. Five states dominate: Haryana, Maharashtra, Karnataka, Andhra Pradesh and Tamil Nadu. The NCT of Delhi shows a high concentration of new firms.

In our prior, Gujarat and Rajasthan might have fared well, but they do not. We may have expected Uttar Pradesh to be weak, but it is similar to Gujarat. West Bengal has a very low density of new firms, showing that conditions there for firm formation have deteriorated when compared with the past.


The authors are researchers at IGIDR. They acknowledge useful discussions with Anjali Sharma.

India's Visa Policy Reforms

by Natasha Agarwal.

In 2016, global travel and tourism contributed US$7.6 trillion to world GDP and supported 292 million jobs worldwide (WTTC, 2017). Undeniably, the sector benefits the economy. It generates employment opportunities and export revenues, creates sectoral linkages, and stimulates infrastructure development. However, to fully reap the benefits from international tourism, countries have to make it easier for travellers to visit. Travellers perceive visa formalities as a travel cost – direct in terms of monetary, and indirect in terms of time spent in waiting in lines, complexity of the process – which exceeding a threshold, can put them off a particular destination or lead them to choosing alternative destinations with less hassle (UNWTO, 2016).

Countries are, therefore, increasingly focussing on visa policies and procedures which in recent years have resulted in some notable progress. In 2015, 61 per cent of the world's population required a traditional visa from the embassy prior to departure, down from 77 per cent in 2008 (UNWTO, 2016). Moreover, between 2010 and 2015, a total of 54 destinations significantly facilitated the visa process for citizens of 30 or more countries by changing their visa policies from "traditional visa" to either "eVisa", "visa on arrival" or "no visa required" (UNWTO, 2016). Ranked at 50, India was identified as one of the 54 destinations that has carried out substantial visa policy reforms.

India's eVisa programme

On 1^st January 2010, India introduced a visa on arrival programme. This was replaced by an eVisa programme on 27^th November 2014. On 1^st March 2016, the visa on arrival programme was re-introduced for nationals from Japan only.

India's eVisa programme allows foreign travellers to apply for their Indian visa online. Once the application is approved, travellers receive an Electronic Travel Authorization (ETA) by an email. The ETA permits their travel to India which has to be within the period mentioned on the ETA. On arrival at the port of entry where the eVisa facility is available, travellers have to present their printed ETA to the immigration authorities who would then stamp the travellers passport thereby permitting entry into the country. Travellers can then travel within India up until the expiry date of the stamped visa in their passport.

Since inception, the program has been reviewed and modified frequently. It is now available to citizens of over 150 countries. The fees are based on principle of reciprocity and categorised into four slaps: 0, 25, 48, and 60 US dollars. There is a bank charge of 2.5%. eVisa’s are categorised into three groups: e-tourist, e-business and e-medical. Double entry is permitted on an e-tourist and e-business visa and triple entry is permitted on an e-medical visa. All three are issued for a period of 60 days, and can be availed up to two times in a calendar year.

Despite these efforts, the data shows a lukewarm response. In year 1 (December 2014 – November 2015), only 5% of international travellers availed their Indian visa online. This fraction grew to 12% in year 2 (December 2015 – November 2016) and just 16% by year 3 (December 2016 – November 2017).

Limitations of the eVisa programme

What explains the low uptake of India's eVisa programme?

A closer examination reveals that the implementation of the programme has been poor.

eVisa's are available under three sub-categories: e-tourist, e-business and e-medical visas. The eVisa online application form lists all the activities which travellers can carry out under each of the three eVisa categories. In addition to choosing a primary purpose of visit, the online application form seems to suggest that travellers are permitted to undertake multiple activities within a group. Upon selecting the visa type on the application form, a pop-up window states "all the following activities are permitted, however select the primarily purpose from the following". This means that a traveller on an e-business visa whose selected primary objective is to, for example, recruit manpower, can undertake any other activity permitted on an e-business visa.

However, instruction number one on "Instructions for Applicant" on the official website states: "e-visa has 3 sub-categories, i.e. e-Tourist visa, e-Business visa and e-Medical visa. A foreigner will be permitted to club these categories (emphasis added)." The online application form therefore permits travellers to combine one activity from each of the three eVisa categories. When a combination of activities is chosen, it is not clear what category of visa will be issued.

Let's imagine a Mr. Smith, a national from one of the eVisa programme eligible countries. While filling his eVisa application form, Mr. Smith opts to primarily participate in a short-term yoga programme (activity listed under e-tourist visa category), attend a business meeting (activity listed under e-business visa category), and go through a short-term medical treatment (activity listed under e-medical visa category). India is liberal: Mr. Smith is not restricted to his chosen primary activity. Nonetheless, the question that arises is: since he has opted for one activity listed under each of the three visa categories, will he travel to India on an e-tourist visa and/or an e-medical visa and/or an e-business visa? Surely, the Indian government does not intend to issue multiple visas to Mr. Smith all in one go.

This is further complicated by the number of entries permitted: double entry on e-tourist and e-business visa, but triple-entry on an e-medical visa. For activities across multiple categories, it is not clear how many entries are permitted. Therefore, will Mr. Smith be given a double or triple entry permit?

An array of questions pertaining to re-entry requirements have also not been answered. For example, would travellers require a re-entry permit if they wish to leave and re-enter India within the 60-day eVisa validity period? If a re-entry permit is required, are there any requirements for issuing this permit, such as a last destination restriction? Do travellers have to re-enter only from the ports of entry at which the eVisa programme is available? Travellers have posed their queries with regards to alternative port for re-entry when travelling on eVisa’s especially when the re-entery is from India’s neighbouring country via land (see here and here).

The programme also has administrative glitches. For example, travellers have repeatedly blogged on the difficulties they experience while using the eVisa application form (see here, here and here). The difficulties of the payment gateway, and limited helpful response from the help desk have been highlighted. Despite the programme being in its third year, these difficulties continue to linger making it an unpleasant experience for travellers.

How can we do better?

Let's run through one problem at a time.

The first area is the permission to come into India under multiple categories. There are two ways to solve this problem.

1. The first solution: change the rules and not permit clubbing activities across the three categories.

   This means that Mr. Smith can primarily choose to either participate in a yoga programme OR attend a business meeting OR a short-term medical treatment of self. For his chosen primary activity, the government would accordingly issue him the visa.

   This is easy to implement. The sentence "A foreigner will be permitted to club these categories." from instruction number one on "Instructions for Applicant" on the official website would need to be deleted. As a result, instruction number one would then read as following "e-visa has 3 sub-categories, i.e. e-Tourist visa, e-Business visa and e-Medical visa. An e-tourist visa, an e-business visa and e-medical would be issued for any activity chosen under the respective visa category." Consequently, the online eVisa application form would also need to incorporate this change, and restrict travellers to choose activities across categories.

   While this is not a great option - we are restricting what visitors can do while travelling, it removes the confusion that has come with clubbing categories.

2. The second and better solution: travellers be permitted to club activities across the three eVisa categories.

   This solution is already in place. However, to avoid the current confusion associated with this solution given the existing premise of the programme, there are two alternatives:

   • Permit three entries under each of the three categories. Once this is done, it is even simpler to abolish the three categories, and just allow anyone to visit India with three entries in 60 days.

   • Permit three entries under each of the three categories. On the eVisa online application form, change e-tourist visa/e-business visa/e-medical visa to tourism purpose/business purpose/medicinal purpose. Accordingly, ETA would reflect the 'purpose of visit' along with the details of the purpose, and the number of entries would be changed to "triple" (see Figure 1 – changes in red). The eVisa stamp in the passport would only state that it is an eVisa, and 'number of entries permitted' would reflect 'triple' (see Figure 1 – changes in red).

     Abolishing the three categories altogether is lucrative. However, having three categories is useful as it facilitates tracking the number of applications within each category which in turn can be used to for drawing sectoral development strategies.

     

   Figure 1: Proposed changes to ETA and eVisa stamp in passport marked in red. Photo courtesy: Traveller's Website

The second area for simplification is the problem of re-entry. This can be solved if the government says:

1. Irrespective of travellers chosen activity across tourist, business and medical categories, the following is applicable:

   • Travellers can re-enter India without a re-entry permit within the 60-day eVisa validity period.

   • There are no last destination restrictions imposed on travellers when re-entering India on an eVisa.

   • Travellers on an eVisa can re-enter India through any port of entry, be it land, air or sea ports

A third improvement lies in greater flexibility for renewing an eVisa in India, capping the number of days a traveller can stay in India on an eVisa. This has been done by many countries e.g. Kenya where travellers can renew their eVisa for a further 90 days at the immigration headquarters in Nairobi, capping the maximum number of days on an eVisa at 6 months.

A fourth area for progress is on information access. The 'Frequently Asked Questions Relating to Tourist Visa' on the website of the Bureau of Immigration, Ministry of Home Affairs, needs an additional FAQ item stating:

1. Travellers who wish to travel to India for a short period, can also avail their India visa through the eVisa programme. Please check the website of the programme (https://indianvisaonline.gov.in/evisa/tvoa.html) for eligibility and requirements.

2. Travellers on an eVisa can carry out activities that spread tourism, medical and business categories. For a list of activities please visit the eVisa online application form.

These four initiatives would solve the flaws of the Indian eVisa program in the following ways. First, it would help in avoiding confusion at the border especially when travellers want to enter and exit India, a country incorporated in intraregional travel plans. Second, it would also help in avoiding policy glitches with respect to the number of entries permitted, and the type of visa issued when activities are chosen across tourism, business and medical categories. Third, it would enable travellers to undertake multiple activities without being held up for violating visa norms. Fourth, it would help rationalise benefits from availing eVisa's over the traditional embassy route visa especially for travellers from countries to whom the government, with effect from 1^st April 2017, provides multiple entry tourist and business visa. This benefit becomes even more apparent when travellers incorporate India in their regional travel plans as travellers have blogged of their preference to avail a traditional visa rather than have to worry about the nitty-gritty missed details of the eVisa programme (see here). Fifth, it would help resolve the consequent administrative glitches especially those that are seen on the ETA and the eVisa stamp in the passport.

Conclusion

To encourage and reap socio-economic benefits from cross border movements of persons, addressing visa procedural bottlenecks can be as, rather even more, important than liberalising visa policies. As a matter of fact, UNTWO reports that communication around visa policies provided by destinations are among the simplest but least addressed areas of opportunity (UNWTO, 2016). While India’s efforts to liberalize the eVisa programme are to be lauded, it needs to ensure that the programme is free from procedural bottlenecks. Afterall, "easy and hassle-free availability of visas is one of the basic ingredients for attracting foreign tourists" (WTTC, 2012).

References

UNWTO, 2016 Visa Openness Report 2015.

WTTC, 2017 Travel and Tourism Global Economic Impact and Issues 2017.

WTTC, 2012 The Impact of Visa Facilitation on Job Creation in the G20 Economies.

 

Natasha Agarwal is an independent research economist affiliated with Research and Outcome Consortium (R&O).

The disaster at Maruti

The news from Maruti is disgusting. I have been curiously watching how the stock market takes it in:

That Maruti has serious labour problems has been known for a long time. But the brutality that unfolded in recent days was out of the world. It was news. When I read about it on Thursday, it seemed to me that Maruti was facing a Tata Motors style situation: of suffering the fixed cost of closing down the existing plant and relocating to a state with better governance. The costs faced in this would be substantial. In that case, a 6 per cent decline of the stock price seemed pretty modest. I watched the small recovery on Friday with surprise. Surely, the cost and complexity of moving out of Manesar is worse than 5%.

Today, on Monday, the market has shifted from a less sanguine assessment to a 10% drop in the stock price. I wonder if this is new information or a modified judgment about how this will play out. Were the speculators on late Friday evening just wrong, or did some new information break?

Bad macroeconomic outcomes and social stress

I would conjecture that poor macroeconomic performance -- low GDP growth and high inflation -- is correlated with greater stress of this nature. With inflation, the logic is straightforward: The worker who had a nominal wage contract finds the need to renegotiate when the value of the rupee changes. This links back to the earlier discussion here on why solving India's inflation crisis is important. Too often, we in India are cavalier about inflation. But we should see inflation as an acid that corrodes all nominal contracts, whether stated or unstated. Renegotiation is costly.

Turning to GDP growth, most people that I know seem to think that a couple of per cent of real per capita GDP growth is important for keeping the peace. A lot of people become a lot more unhappy when growth slows. Indian democracy does a pretty good job of containing the angst. There will be no revolution here. But life is substantially easier if the engine of GDP growth is purring. When it stalls -- as it appears to have done in 2012 -- a whole host of social problems erupt.

Law and order as the fundamental foundation of civilisation

This is a reminder to us about how law and order is the fundamental precondition of civilisation. The most important public good of all, the first claim on the resources of the State including the time and attention of the senior leadership, is police, courts and laws. The entire story of the market economy and high GDP growth can only come about when safety of life and property is guaranteed. The events in Maruti are an important reminder to every investor about the weaknesses of governance in Haryana.

In tracking conditions in any state, I find it useful to watch the time-series of the share of the state in the overall all-India investments outstanding, that are `under implementation', in the CMIE Capex database. Here's an example, for Bihar:



November 2005 is the date that Nitish Kumar became the new CM of Bihar. He is widely reputed to have made important progress on improving law and order. At first, the share of Bihar (in all-India under implementation investment) continued to drop. I am sure the changes brought about by Nitish took time; Rome wasn't built in a day. And, after improvements come about, skeptical investors would take some time in making up their minds that conditions are now better. From 2009 onwards, it appears that there is some upward movement. The overall gain seems to be roughly 1 per cent of the all-India total, which is a significant change.

Compare this with Haryana:

There was a big spurt in the share of Haryana in the overall under-implementation investment in India. After that, the numbers have steadily trended down. Is Haryana suffering from a resource curse in terms of proximity to Delhi?

Rethinking labour law

In the early decades about independence, India constructed a remarkable legal framework which was strongly pro-trade-union. Few countries have enshrined trade unions into laws on the scale that India has done. In those years, trade unions were primarily led by socialist/communist parties. While we may disagree with their views, there was a fundamental decency about them. Some of the best human beings in India, in the 1950s and 1960s, were communist. Perhaps this coloured our thinking, and encouraged us to respect and empower trade unions strongly in the legal framework which fell into place over the 1960s and the dark days of the 1970s.

Today, a hyper-empowered trade union is a potent tool for extortion in the hands of local goons. To solve this problem, it is important to rethink the checks and balances embedded in labour law, which have gone too far in the direction of making trade unions strong. Now that we know that the people in trade unions are most likely local goons, do we want to hyper-empower them through labour law?

Addressing the problems of Rupeezone

While everyone is pondering the ways in which Eurozone is not an optimal currency area, I found myself worrying about the ways in which Rupeezone is not an optimal currency area. In the Financial Express today, I have a column: Addressing the problems of Rupeezone.

Here are interesting materials on Greece which set the stage for this:

• A money too far by Paul Krugman, in the New York Times.
  
• A response by Greg Mankiw on his blog. He refers to this AER article from 1986 by Christine Romer, and this article by Christopher Hanes in AER in 1993.
  
• A response by Paul Krugman on his blog.
  
• A blog post by David Beckworth.
  
• On the subject of labour mobility, Casper van Ewijk and Michiel van Leuvensteijn have an important column on voxEU making the link between labour market mobility and the transactions taxes which governments impose on real estate transactions.
  
• Here is a column by Feldstein on these issues back in 1998.
• Greek lessons for India, in the Business Standard, by Suman Bery.

This is an interesting demo of how economics happens today: an interleaving between journal articles, newspaper columns and blog posts.

Investment activity at the state level

Ila Patnaik looks at what is happening in the CMIE Capex database on investment in Gujarat, Maharashtra, West Bengal, Andhra Pradesh and Orissa. The measure of interest is: the time-series of the share of the state in overall projects `under implementation' in the CMIE database. Each of these states holds an interesting story.

Also see Kunal Sen in Financial Express on the evolution of governance at the state level in India.

What if India had a Hong Kong?

An alternate history that I find interesting is a scenario where, in 1947, the British kept one city in India - e.g. Surat. This is analogous to the British control of Hong Kong in China after the communist revolution.

Why is Surat interesting for such an analysis, and not Bombay? It sounds too implausible for free India to have tolerated colonial rule for Bombay. A solution like Hong Kong, Macao or Goa for a less important place is more plausible. The other reason why Surat is of interest is that before Shivaji sent the merchants of Surat scurrying to the safety of British-controlled Bombay, Surat was the commercial capital of the West Coast. So there is perhaps some natural geographical advantage of that location.

If the British had run Surat in the fashion that they did for Hong Kong, how might this have changed India's trajectory? The analogy with Hong Kong is straightforward. In this scenario:

• Surat would have become a place with a market economy, with strong public goods of law and order, judiciary and legal system.
• When India embarked on socialism, this would have been a place for people and capital to go to. Some of the brain drain and capital drain that India suffered to locations all over the world would have instead gone to Surat.
• Surat would have then become a key mechanism for India to plug into globalisation, for trade in goods and services and for financial services.
• When India started stepping out of socialism, a good deal of institutional capability, human capital and financial capital would have been ready at hand to help get the mainland going again.
• When a country wants to undertake institutional reform, it is quite useful to have `regional role models' (a term drawn from the World Bank's East Asian Miracle book). India unfortunately has few regional role models other than the good work done in Sri Lanka on trade liberalisation before the war, and the work done in Bangladesh in microfinance; this is in contrast to East Asia where each country is able to pick and choose from regional success stories in any area of reform. If Surat had been a Hong Kong, then institutional arrangements there would have been a natural starting point for thinking about legal, regulatory and institutional development in India. This would have given faster institutional evolution and thus growth in India, once India wanted to actually do institutional reform.
• The last point is a bit speculative. Suppose Surat was a vibrant outpost of good institutions and laissez faire, while India was headed off into a bad institutions and dirigisme from the late 1950s onwards. Would the very existence of a visible alternative have modified India's trajectory? It is easy to think that from the early 1990s, when India was getting interested in reintegrating into the world economy, and in building institutions, that a Hong Kong would have helped. But look back even before that; would India's long descent have been reduced or even averted by having a counterpoint? We know that in the Chinese case, they had Hong Kong and still suffered from the disasters of the cultural revolution. But in a functioning democracy with freedom of speech, the power of ideas and impact of information is greater.

In summary, if you think that China's incredible economic success was aided by having laissez-faire Hong Kong handy, then in this alternate history, a similar evolution for Surat would have helped India.

I recently came across similar arguments being made by Paul Romer. He uses the term `Bridge Cities' for such cities, which can help speed up the development of the host country.

Investment is destiny

Geographical patterns in investment planned out today are the geographical patterns in per capita GDP of coming years. Ila Patnaik reviews what the CMIE Capex data says about investment in the big states of India. The surprise, for me, was the poor showing of states that one traditionally expects will do well: Tamil Nadu, Karnataka, Andhra Pradesh and Maharashtra (they're at ranks 7-10). At the same time, this is an encouraging phenomenon in that investment isn't all going to rich and successful states.

Measurement of quality of service of electricity distribution

What you measure is what you can manage. In thinking about public goods in India, achieving a sound public policy framework in India, and appropriate checks and balances for providers, critically requires sound measurement of outcomes.

In the area of telecom, TRAI has started working on measurement of quality of service (QOS). As an example, look at the PDF file that they are putting out. I find it interesting, but not yet designed in a way that facilitates customer choice. When you're picking a mobile phone vendor in Delhi or Bombay, it is hard to read this file and understand which vendors are strong on QOS. In the area of electricity, a think tank named Prayas has embarked on measurement of the quality of service at a few locations in Pune. They have setup a few data loggers which generate realtime data for the mains voltage, and measure supply interruptions. They have released documentation, and a lovely google maps mashup which locates their data loggers and gives access to the information being captured (click on any of the balloons). Their email says:

For the first time in India, the ESMI captures supply interruptions data as well as voltage levels at the ordinary consumer location.

Poor supply quality (i.e. frequent supply interruptions and low voltage levels), is the most common complaint by electricity consumers and often results in consumers unwillingness to pay. Poor supply quality forces consumers to either invest in back-up devices such as stabilisers, inverters and generators or they suffer loss of productivity and inconvenience. Though recently efforts are being made to capture supply quality parameters, (e.g. CEA compilation of interruptions at 11 kV feeders in urban areas) there is still a long way to go for availability of reliable and comprehensive indicators of supply quality.

The Electricity Supply Monitoring Initiative (ESMI), by Prayas, is developed as a complement to such ongoing efforts to monitor electricity supply quality. This is a tool for consumers and regulators to get an idea of the ground reality and to increase the accountability of electricity utilities. For example, this will allow consumers to verify if load shedding is being carried out as per the regulatory commission's directive. Similarly ESMI data could be used to assess the impact of large capital expenditure (under projects such as APDRP or MSEDCL's infrastructure plan) on supply quality.

Currently data loggers are installed at three locations in Pune, and supply interruptions and voltage profile of these locations is available at http://prayas.icantrack.com. This site will be updated periodically and many more locations will be added in the next few weeks.

As an example, their Pune results reveal interesting and new facts:

Pune does not face any scheduled load shedding due to implementation of `Pune Model'. Under this arrangement local industries give the required relief to the grid by generating power through their captive power plants (using liquid fuel such as diesel). In spite of this as shown in the tables, Pune consumers fact ten to twenty sustained interruptions (i.e. interruptions of 4 minutes and more) every month and do not have supply for about eleven hours every month. As per Maharashtra Electricity Regulatory Commission (MERC) regulations, voltage for low-tension consumers should be between 225 volts to 255 volts. But as shown in the following tables, voltage levels in Pune are significantly below MERC norms. For example, even in central Pune (Deccan Gymkhana) voltage is often below MERC norms and for about 7% to 9% of the time is very low (i.e. below 195 Volts). This indicates that, apart from sufficient availability of power, proper maintenance and strengthening of local distribution network is essential to ensure good quality of supply. The electricity supply quality in rural areas is expected to be still worse. In the next few weeks ESMI will focus on capturing status of electricity supply quality in rural areas.

I'm a great believer in the importance of information in closing the loop and enabling superior public goods. What is not easy to figure out is the incentive structure for this information. This information has public goods characteristics, so what would work best is for a government agency to contract-out its production and public dissemination with tight integration between the public goods outcomes data and google earth. However, this would require sensible decision making and procurement on the part of the government agency. There is a bit of a chicken-and-egg problem here. A dysfunctional and non-responsive government agency would not even make this first step correctly - it might do nothing, or it might muck up the contract, or the contract might get awarded to the wrong team. Without a feedback loop from outcomes to citizens to policy, these mistakes might not get corrected.

The outlook for Maharashtra and Gujarat

R. N. Bhaskar has written a three-part article in DNA about the decline of Maharashtra. This particularly caught my eye because I just noticed that in the CMIE Capex data, Maharashtra is at rank 10 amongst the 20 large states when ranked by the value of projects `under implementation' expressed on a per capita basis. I was very surprised when Maharashtra came out at the median and not around the top. The top five are Haryana, Orissa, Himachal Pradesh, Gujarat and Uttarakhand.

I have long noticed the theme of `finance follows trade'. Where international trade blossoms, there is a natural consumption of international financial services. Bhaskar rightly points out the edge that has developed with Gujarat in ports. He writes about the remarkable Mundra port. I have long been struck by the fact that Gujarat has roughly half of India's ports. Putting together the picture on roads, ports, and the possible freight corridor, this could give Gujarat a very good position on international trade.

So could Gujarat make a play for doing international financial services? The MIFC report is not deeply bound to Bombay: it focuses on getting the framework of financial and monetary policy in order, so that India can produce IFS. While Bombay has a natural edge, it doesn't have to be the case that an IFC takes root in Bombay. After all, only a few hundred years ago, Surat was a big commercial centre and Bombay was a barren island. :-) Bhaskar talks about the work that is going on at the outskirts of Ahmedabad to will a finance cluster named `GIFT' into existence. GIFT has a new website. Update: Some people in government seem to think similarly about the lack of binding to `M' in the MIFC report.

How states are faring on investment

Ila Patnaik summarises the CMIE Capex data on investment projects at hand in the states of India. The ranking, by the value of `under implementation' projects expressed per capita, is most interesting. It makes us think afresh, and challenge many preconceptions, about which states are doing well and which states aren't.

A surprising feature of the data is the remarkably poor showing of Maharashtra. This is a failure of the politicians. As Sunil Sethi says in Business Standard today:

In its edgy, manic way, Mumbai prided itself on being the more organised of Indian cities, devoted to the work ethic, the go-getting glamour capital of the country. This is no longer true. It is now more ramshackle than Kolkata, more inefficient than Delhi and, probably, neither as rich nor as inventive as Bangalore or Hyderabad. Its cosmopolitan ethos and egalitarian energy has been hobbled by provincial-minded politicians, sectarian ideology and pick-pocket capitalism.

Should IST be broken up into two? Should India do DST?

Evaluating two timezones and Daylight Saving Time for India, by Viral Shah & Vikram Aggarwal, August 2007. On a related subject, you might like to see this blog entry. Update: Pierre Mario Fitter has an article in Business World on this question.

Good effort on combining maps with economic data

In the EPW of 22 April 2006, I wrote this book review of Social & Economic Profile of India by Peeyush Bajpai, Laveesh Bhandari and Aali Sinha. It is published by Social Science Press, 2005.

In the old days, there were the fascinating established disciplines of `regional economics' or `economic geography'. These lost ground over the decades, and well-trained contemporary economists have typically not studied these fields. The typical modern economist is highly conscious of time-series econometrics, but has only a rudimentary grasp of handling spatially organised data. In recent years, however, economic geography has come to enjoy a renaissance, through three factors.

The first is the new work by Paul Krugman and others which seeks to link up geography into the core questions of economics. For example, while classical trade theory focused on gains from trade through differences in factor endowments, contemporary trade theory is greatly concerned with geographical distance. The second factor at work is the heightened interest in spatial distributions of income and purchasing power on the part of the marketing fraternity, which has brought new kind of interest in learning about these questions as also an impetus for improving the informational foundations. The third factor at work has been the remarkable marriage of computers and maps with economic data, which has made it easy to visualise spatial data using interactive software systems which go by the fancy name of `geographical information systems' (GIS).

Bajpai, Bhandari and Sinha have come out with a volume titled `Social & Economic Profile of India'. In my knowledge, this represents a first effort at marrying computers, maps and economic data to produce an economic atlas of this fashion. So far in India, there are some strong GIS systems, and there are strong spatial databases. This book is the first time these have come together in a satisfactory fashion.

The authors must be commended for scouring the statistical system for spatial data. In some cases, the unit of observation is the state. In some cases, the unit of observation is the district. A very wide range of sources have been tapped to put together this volume. In some cases, the `standard sources' do not offer a certain kind of information, but the authors have reduced NSSO or NFHS data into summary statistics organised by location.

Every page of the volume is a map (in colour). Social Science Press has done an equally remarkable job of bringing world-class paper and printing to bear on this problem. As little as five to ten years ago, it was not possible to envision a book like this about India, but this has now become a reality.

The content of the book is organised as sections on Demography, Geography, Poverty and hunger, Health, Education, Water and sanitation, Employment, Fiscal analysis, Mass media, Safety and justice, Economic profile and Decentralisation. This is a fair depiction of what the Indian statistical system offers in the field. I was particularly impressed at the effort that the authors have taken to obtain data on issues of law and order and the judiciary, which tend to be ignored by the economics profession. Problems of law and order are increasingly shaping up as a central task of the State in addressing the poverty traps in the country, since participation in markets, investments in human and physical capital, and the equalising-differences of the price system cannot come into play until safety of life and property is assured.

I wandered through the book from end to end several times. Such such unstructured browsing is strongly recommended for the intelligent layman, and I am sure that each reader will take away different insights from this material. If there is one thing that I am struck by, it is the extent to which the stories are correlated. There is a powerful single factor called economic growth, which appears to affect a diverse array of spatially organised data.

Great essay on globalisation

I loved the book From Beirut to Jerusalem by Thomas Friedman (1998) when I read it (at the time). After that, I enthusiastically bought and read his The lexus and the olive tree. I got something from that book but not a whole lot. When The world is flat came out, I read an extended essay which he put out in the The New York Times and chose to skip the book. Friedman engages in too much rhetoric and now I see the book as the latest `management guru' book. I never read management gurus.

Today, I read a marvellous essay on the book by Edward Leamer which is forthcoming in Journal of Economic Literature. Much unlike Thomas Friedman, it is actually insightful and (atleast for me) said new things about globalisation other than the obvious insights of gains from trade based on diminishing frictions. If you've been uptodate with the field of international trade, it might be familiar territory, but for the rest, it's a great read.

I was struck by the fact that over time, the distance coefficient in gravity models has not dropped. It is a really surprising result. I wonder why this happens. Certainly, looking from an Indian perspective, there has been a huge upsurge of trade to faraway locations, made possible by reduced transaction costs. In particular, in terms of BPO, India appears to do huge trade with faraway locations (e.g. UK and UK) and negligible trade with close-by places. This doesn't sound like a gravity model story to me. (Richard Portes & Helene Rey have this famous result from 1999 on gravity model type stories with capital flows also, and there also, I feel that India's a bit of an outlier, doing very little business with the immediate neighbourhood.)

One possibility that struck me was like this. In the last 30 years, there have been important political developments leading to lowered trade barriers for intra-regional trade in 3 regions: EU, North America and (to a lesser extent) East Asia. Each of these events - e.g. NAFTA - might give an upsurge of trade between countries at short distances to each other, thus swamping the effects of improvements in the transportation of goods. If this happened, then it's a one-off, and in the future, we may get a decline in the distance coefficient.