Abstract
This chapter seeks to provide some indicative applications of the unidimensional development ranking method suggested in the previous chapter. For an application to the problem of studying time trends, it considers the Indian economy in the recent decades and studies (separately) the cases in which consumer expenditure (as proxy for income) and wealth are the relevant dimensions of development. As a cross-sectional application, it compares the levels of development of the BRICS countries in a recent year (2017), considering wealth to be the dimension of interest. So far as the data source for consumer expenditure is concerned, we follow the standard practice of using the findings of the quinquennial large-sample surveys conducted by the National Sample Survey Office. The All-India Debt and Investment Surveys constitute the data source for wealth distribution in India. For the cross-sectional exercise on the BRICs countries in 2017, the Global Wealth Report for that year by Credit Suisse has been used. It is found that our suggested methodology is a non-trivial extension of the conventional crisp (i.e. non-fuzzy) theory in the sense that in a number of cases in which the conventional approach fails to rank two economies (or the same economy at two points of time) in terms of the level of development, it does yield definitive conclusions. Since we use real (rather than hypothetical) data, these findings seem to provide support to the suggested methodology.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsNotes
- 1.
Information regarding aggregative magnitudes such as per capita income or wealth can be sought from macroeconomic data. However, since our indices are distribution-sensitive, we need information on how the aggregative figures are distributed among the population.
- 2.
The two Indian Human Development Surveys (IHDS) conducted by the National Council of Applied Economic Research (NCAER) in collaboration with the University of Maryland in 2004–05 and 2011–12 are perhaps the only sources of good-quality data on income distribution in India. (On the 2004–05 study, see Desai et al. (2010); on 2011–12 see Desai and Vanneman (2018).) These surveys also have a lot of information on a number of other important socio-economic variables. However, the data generated by these studies would only enable us to carry out a ranking of the levels of development (as measured by income) in these two years.
- 3.
In between these five-yearly large-sample studies, the NSSO also conducts thin sample studies from time to time. However, because of the smallness of the samples, these studies are usually not subjected to detailed analysis by academic researchers.
- 4.
Calculation of the all-India Lorenz curves often requires going to the unit-level data generated by the surveys because the published official reports often give only the curves (or, at least, the information necessary for calculating them) for the rural and the urban sectors separately. The all-India average MPCE for any year can, however, be calculated relatively easily from the published figures on sectoral averages given in the reports by taking their population-share-weighted arithmetic mean.
- 5.
The questionnaires were printed with the list of goods down the left-most column. The next two columns elicited information on quantities and expenditure over the last seven days. The last two columns asked for information on the same points for the last 30 days.
- 6.
It may seem that since data were collected from each household on both high- and low-frequency items with the 30-day recall period, these could easily be used to deduce total expenditure with this recall period and the resulting estimates can be compared with those for the other large-sample-survey years. The additional data on high-frequency items with the seven-day recall period can just be ignored for the purpose of this exercise. However, one major problem here is that when the respondents are asked to report on expenditure on any item, they are effectively prodded to reconcile their rates of consumption across the two periods. This affects data quality. Deaton (2003) notes that there is some evidence that is consistent with this type of reconciliation.
- 7.
In the survey report, MPCE meant the MPCE based on the 30-day recall period although it was also sometimes called “MPCE(U30)” or “unadjusted MPCE”. In contrast, the MPCE that used the 365-day-recall-period data on the five low-frequency items mentioned in the text was called “MPCE(M)” or “adjusted MPCE”. It is conceivable that the problem of the type mentioned in Note 6 above might persist here to some extent. However, the problem would be much less pronounced here because the items involved are now the infrequently purchased items (unlike in the 1999–2000 survey when it was the frequently purchased items on which such dual information was elicited). It may reasonably be assumed that problem of reconciliation would be less severe here.
- 8.
Again, theoretically, there is here the possibility of persistence of the Deaton-type problem since, although the two types of schedules were used on two different sets of households, the households interviewed with Schedule Type 1 were asked for information regarding a subset (Category I) of the items on the basis of both the 30-day and the 365-day recall periods. However, since these were the relatively low-frequency items, remarks similar to those in Note 7 apply. It can be assumed that the problem is of negligible proportions.
- 9.
If there is a similar trend in the degree of inequality between 1983 and 1993–94, it is possible to argue that the trend between 1993–94 and 2004–05 is a continuation of a longer-term secular increase in inequality of the MPCE distribution in the country and that the matter is unrelated to the reforms of the 1990s.
- 10.
The conditions on F mentioned in the text are natural requirements of a fuzzy measure of development. The matter was discussed in Chap. 3.
- 11.
In the text, we went into some but not all of the (cardinal) issues regarding the magnitudes of the increase in average MPCE over the period and in the different subperiods. In a cardinal framework, it would be of interest to note that in the decade following the reforms of the early 1990s, the index value in Table 4.5 changed from 112.73 in 1993–94 to 125.51 in 2004–05, registering an 11.34% increase. In the decade preceding the reforms, the change was from 100 in 1983 to 112.73, posting a 12.73% increase. Thus, there was actually a slight decline in the growth rate of average MPCE in India in post-reforms decade. However, the growth record of the pre-reforms years is also none too impressive. In fact, in the six years between 1987–88 and 1993–94, there was paltry 2.46% increase. Also, the growth record has improved in more recent times. Between 2004–05 and 2011–12, the rate of growth was about 28%.
- 12.
The question here is whether it is the case that an individual’s rank in consumer expenditure distribution is the same as the rank in the wealth distribution. The authors demonstrated that this was far from being true either in 1991 or in 2002.
- 13.
For a discussion of the data limitations in this context see Subramanian and Jayaraj (2006).
- 14.
Information was solicited on wealth at a specified point of time in the preceding years. Thus, the surveys conducted in 1992, 2003 and 2013 generated data on assets held by households (at some points of time) in 1991, 2002 and 2012, respectively.
- 15.
There does exist price indices for some types of financial assets such as equity. However, such assets constitute a small percentage of total assets in India.
- 16.
Zacharias and Vakulabharanam (2009) use the Consumer Price Index for Agricultural Labourers for deflating nominal rural wealth and that for industrial workers for deflating nominal urban wealth
- 17.
The sectoral population shares needed for this exercise for the year 2002 was calculated by using interpolation based on the 2001 and the 2011 Censuses. The procedure was similar to that behind Table 4.4. The needed figures for 2012 were based on extrapolation.
- 18.
We have been encouraged to construct our notional combined CPI not only by the fact (mentioned in the text) that its underlying idea seems to be the same as the one behind the official new series with 2012 as base year but also by the observation that the results of our exercise seem to tally with the information given in the long time series on “CPI inflation in India” given on several unofficial websites such as www.inflation.eu, www.calculatorstack.com, www.globalrates.com etc. None of these, however, seems to explain how these figures were arrived at in the absence of an official combined CPI series for the pre-2011 years.
- 19.
An implicit assumption in the argument stated in the text is that in any given year everybody faces the same prices. If it is the case that individuals in different quintiles face different prices and if the pattern of these differences changes when we move from one price deflator to another, the argument will cease to be valid. The assumption of identical prices, however, has always been made in this type of empirical research, at least in the Indian context.
- 20.
Actually, the figures in Table 4.9 were calculated from the distribution of nominal wealth. However, in view of the discussion in the preceding paragraph in the text, the results of the calculations would be the same if we express the figures in constant prices.
- 21.
For the Indian case, the net worth distributions can, in principle, be calculated for all the three years 1991, 2002 and 2012 referred to in our discussion above since the relevant data were collected at the unit (household) level. However, this information at the all-India level (combining the information for the rural and the urban sectors) for the year 1991 does not seem to be readily available in the reports. (While the mean net worth level is easily obtainable, we also need to know the pattern of its distribution.)
- 22.
It may be noted, however, that the practice of computing the level and the distribution of wealth per capita (followed in Tables 4.8 and 4.9 in the preceding subsection in the text) also has its own uses in the context of countries like India. In these countries, the role of wealth in guarding against economic and other vulnerabilities of all members of the households (including children) is one of the main reasons for studying the pattern of wealth ownership (in addition to that of current income or expenditure).
- 23.
Conventionally, Lorenz curves are defined for distributions of non-negative variables. However, if the mean value of the variable is positive, the extension to the case of variables that can take negative values at some points is a simple matter. The definition of the Lorenz curve remains exactly analogous. Since, trivially, the shares of the bottom 0% and the bottom 100% of the units of observation are 0% and 100%, respectively, the curve starts, as usual, at (0, 0) and ends at (100, 100). The egalitarian line remains the same. The only difference from the usual case is that some points on the Lorenz curve may now have negative ordinates. However, that makes no difference to the notion of Lorenz dominance: again, a distribution x strictly Lorenz dominates a distribution y if and only if the Lorenz curve of x is nowhere below that of y and is above it at some points. Moreover, our criterion of fuzzy Lorenz dominance applies: x Lorenz dominates y in the extended (fuzzy) sense if and only if N(x, y) > N(y, x) irrespective of whether the Lorenz curves intersect or not.
References
Anand I, Thampi A (2016) Recent trends in wealth inequality in India. Econ Polit Wkly 51(50):59–67
Credit Suisse (2017) Global wealth data book 2017. Credit suisse research institute
Deaton A (2003) Adjusted Indian poverty estimates for 1999–2000. Econ Political Wkly 38(4):322–326
Desai S, Dubey A, Joshi BL, Sen M, Shariff A, Vanneman R (2010) Human development in India: challenges for a society in transition. Oxford University Press, New Delhi
Desai S, Vanneman R (2018) Indian human development survey-II. Inter-University consortium for political and social research, Ann Arbor, MI
Jayadev A, Motiram S, Vakulabharanam V (2007) Patterns of wealth disparities in India during the liberalisation era. Econ Political Wkly 42(38):3853–3863
NSSO (1998) Debt and investment survey. NSS Forty-Eighth Round. National Sample Survey Organisation, Department of Statistics, Government of India
NSSO (2006a) Household asset holdings, indebtedness, current borrowings and repayments of social groups in India. NSS 59-th Round. National sample survey organisation, Ministry of statistics and programme implementation, government of India
NSSO (2006b) Level and pattern of consumer expenditure 2004–05. NSS 61-st Round. National sample survey organisation, Ministry of statistics and programme implementation, Government of India
NSSO (2013) Key indicators of household consumer expenditure in India. NSS 68-th Round. National sample survey office, Ministry of statistics and programme implementation, government of India
NSSO (2014) Key Indicators of debt and investment in India. NSS 70-th Round. National sample survey office, Ministry of statistics and programme implementation, government of India
Pal P, Ghosh J (2007) Inequality in India: a survey of recent trends. In: Jomo KS, Baudot J (eds) Flat world, big gaps: economic liberalization, globalization, poverty and inequality. UN Publications, New York
Shorrocks AF (1983) Ranking income distributions. Economica 50(197):3–17
Subramanian S, Jayaraj D (2006) The distribution of household wealth in India. Research Paper No. 2006/116. UNU-WIDER, United Nations University
Zacharias A, Vakulabharanam V (2009) Caste and wealth inequality in India. Working Paper No. 566. The Levy Economics Institute of Bards College
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
Copyright information
© 2020 The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this chapter
Cite this chapter
Banerjee, A.K. (2020). Illustrative Applications of Unidimensional Development Indices. In: Measuring Development. Themes in Economics. Springer, Singapore. https://doi.org/10.1007/978-981-15-6161-0_4
Download citation
DOI: https://doi.org/10.1007/978-981-15-6161-0_4
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-15-6160-3
Online ISBN: 978-981-15-6161-0
eBook Packages: Economics and FinanceEconomics and Finance (R0)