Jul 25 2014 : The Economic Times (Delhi)
POINTS TO PONDER - Let There Be Light
Bibek Debroy
|
Being lit-up may be the brightest index for measuring development
Unless your general know ledge is particularly good, you may not have heard of Gbadolite, as opposed to Kinshasa, capital of the Democratic Republic of the Congo (DRC). Gbadolite is the capital of the country's Nord-Ubangi district.More importantly , it was the ancestral home and residence of ex-President Mobutu Sese Seko, president of Zaire (DRC's former name), between 1965 and 1997.
Mobutu wanted to change Gbadolite into a “Versailles“ in the jungle.
For years, economists have sought measures of economic development, per-capita GDP not being the only option. GDP is market value of goods and services produced in a year and per-capita GDP needn't be the same as per-capita income (income includes net factor payments). For instance, for Indian states too, if remittance income is significant, per-capita state gross domestic product (GSDP) can deviate significantly from a state's per-capita income.
`Gross' Domestic Product As a measure, GDP is often criticised. The thrust of that criticism is an inability to quantify variables that don't have markets andor prices.
None of that criticism is really new.
People who point to these warts often don't know that Simon Kuznets, father of national income measurement, wrote one of the best critiques of GDP in 1934. However, income is, at best, a means. It isn't an end. We must have some other measure of human development and deprivation.
Consequently , we went through stuff like quality-of-life indices and the most robust of measures now is UNDP's human development index (HDI, based on purchasing power parity , per-capita income, health and education). Every once in a while, someone comes up with a new measure, incorporating new ideas and variables. But in considering development or deprivation, several variables are correlated with several others.
The Cows Come Home Idon't see much point in adding to complexity . Taken a bit out of context, this is a bit like Occam's Razor: “Pluralitas non est ponenda sine necessitate.“ Simplified and paraphrased: don't complicate without reason.
Down the years, Laveesh Bhandari and I have done some work on interstate comparisons.
Once fishing through livestock census data, we found some rather unexpected correlations. Whenever a state's agriculture did well, there seemed to be reverse gender preference. Usually , the ratio of cowsbulls was 50:50. However, whenever agriculture did well, the ratio became more like 70:30. Once you think about it, the finding makes sense.
Agriculture doing well means mechanisation (less bulls needed) or diversification into dairying (less bulls wanted). Ditto for dogs, where the state of data is much more unsatisfactory . Whenever a state does well, “pet“ dogs increase relative to “non-pets“, a phenomenon also linked to urbanisation. Had data permitted, one would probably have found a male dog preference, correlated with urbanisation. We could, thus, have developed cow and dog indices for development.
We didn't do that because we didn't want to become the butt of jokes.
But there is one other thing we fou nd. Through Nasa, we got hold of pic tures of India taken at night. There was a very visible correlation bet ween darkness and states known to be backward, and illumination and states known to be more advanced.
For instance, central and eastern In dia was shrouded in relative dark ness. You could see that pattern glob ally too, with highways lighting up parts of the US.
Enlightened Development Our exercise was largely fun. Now, there is a serious paper by Paul Ras chky (Monash University) and Ro land Hodler (University of St Gallen) published in The Quarterly Journal of Economics, illuminating the dism al science. What's remarkable is the quality of satellite data they have col lected: 38,427 sub-national regions, between 1992 and 2009. Therefore, you not only have a light shot at one point in time, you also have a time se ries. One hypothesis is obvious, such as North Korea versus South Korea, or the Nile region in Egypt. Deve lopment means light, deprivation means darkness.
However, because of time series, the fascinating story lies elsewhere.
Which is where Gbadolite comes in.
When a political leader comes to power, his favoured geographical re gion lights up, such as Gbadolite un der Mobuto. To make the story plausible, after Mobuto's exile and death, Gbadolite faded again.
There is a similar story -without the fade-out bit -for Hambantota in Sri Lanka under President Mahinda Rajapaksa. Subject to quantification and estimation problems, here are some figures.
If literacy is high and democracy mature, light intensity doesn't alter with change in political leadership.
But if these indicators are “medium“, night-time light intensity in the leader's birthplace (or favoured region) increases by 4% and regional GDP by 1%. If these indicators are “poor“, those numbers change to a light intensity increase of 11% and regional GDP increase of 3%.
It isn't obvious from the paper whether these consequences follow from new resources or diversion of resources from other regions. Nevertheless, it is a novel study. Lenin had a definition of communism as Soviet power plus electrification. If one can ensure 24×7 electricity to all households -not just token connections at the village-level and villages that are off-grid -socioeconomic development becomes a given. Clearly, we picked up the wrong kind of “power“ from Lenin's definition.
The writer is consulting editor, ET
Night pictures taken by Nasa show a visible correlation between darkness and states known to be backward: central and eastern India was under relative darkness
Mobutu wanted to change Gbadolite into a “Versailles“ in the jungle.
For years, economists have sought measures of economic development, per-capita GDP not being the only option. GDP is market value of goods and services produced in a year and per-capita GDP needn't be the same as per-capita income (income includes net factor payments). For instance, for Indian states too, if remittance income is significant, per-capita state gross domestic product (GSDP) can deviate significantly from a state's per-capita income.
`Gross' Domestic Product As a measure, GDP is often criticised. The thrust of that criticism is an inability to quantify variables that don't have markets andor prices.
None of that criticism is really new.
People who point to these warts often don't know that Simon Kuznets, father of national income measurement, wrote one of the best critiques of GDP in 1934. However, income is, at best, a means. It isn't an end. We must have some other measure of human development and deprivation.
Consequently , we went through stuff like quality-of-life indices and the most robust of measures now is UNDP's human development index (HDI, based on purchasing power parity , per-capita income, health and education). Every once in a while, someone comes up with a new measure, incorporating new ideas and variables. But in considering development or deprivation, several variables are correlated with several others.
The Cows Come Home Idon't see much point in adding to complexity . Taken a bit out of context, this is a bit like Occam's Razor: “Pluralitas non est ponenda sine necessitate.“ Simplified and paraphrased: don't complicate without reason.
Down the years, Laveesh Bhandari and I have done some work on interstate comparisons.
Once fishing through livestock census data, we found some rather unexpected correlations. Whenever a state's agriculture did well, there seemed to be reverse gender preference. Usually , the ratio of cowsbulls was 50:50. However, whenever agriculture did well, the ratio became more like 70:30. Once you think about it, the finding makes sense.
Agriculture doing well means mechanisation (less bulls needed) or diversification into dairying (less bulls wanted). Ditto for dogs, where the state of data is much more unsatisfactory . Whenever a state does well, “pet“ dogs increase relative to “non-pets“, a phenomenon also linked to urbanisation. Had data permitted, one would probably have found a male dog preference, correlated with urbanisation. We could, thus, have developed cow and dog indices for development.
We didn't do that because we didn't want to become the butt of jokes.
But there is one other thing we fou nd. Through Nasa, we got hold of pic tures of India taken at night. There was a very visible correlation bet ween darkness and states known to be backward, and illumination and states known to be more advanced.
For instance, central and eastern In dia was shrouded in relative dark ness. You could see that pattern glob ally too, with highways lighting up parts of the US.
Enlightened Development Our exercise was largely fun. Now, there is a serious paper by Paul Ras chky (Monash University) and Ro land Hodler (University of St Gallen) published in The Quarterly Journal of Economics, illuminating the dism al science. What's remarkable is the quality of satellite data they have col lected: 38,427 sub-national regions, between 1992 and 2009. Therefore, you not only have a light shot at one point in time, you also have a time se ries. One hypothesis is obvious, such as North Korea versus South Korea, or the Nile region in Egypt. Deve lopment means light, deprivation means darkness.
However, because of time series, the fascinating story lies elsewhere.
Which is where Gbadolite comes in.
When a political leader comes to power, his favoured geographical re gion lights up, such as Gbadolite un der Mobuto. To make the story plausible, after Mobuto's exile and death, Gbadolite faded again.
There is a similar story -without the fade-out bit -for Hambantota in Sri Lanka under President Mahinda Rajapaksa. Subject to quantification and estimation problems, here are some figures.
If literacy is high and democracy mature, light intensity doesn't alter with change in political leadership.
But if these indicators are “medium“, night-time light intensity in the leader's birthplace (or favoured region) increases by 4% and regional GDP by 1%. If these indicators are “poor“, those numbers change to a light intensity increase of 11% and regional GDP increase of 3%.
It isn't obvious from the paper whether these consequences follow from new resources or diversion of resources from other regions. Nevertheless, it is a novel study. Lenin had a definition of communism as Soviet power plus electrification. If one can ensure 24×7 electricity to all households -not just token connections at the village-level and villages that are off-grid -socioeconomic development becomes a given. Clearly, we picked up the wrong kind of “power“ from Lenin's definition.
The writer is consulting editor, ET
Night pictures taken by Nasa show a visible correlation between darkness and states known to be backward: central and eastern India was under relative darkness
