Are the world’s poorest being left behind?


Traditional assessments of economic growth and progress against poverty tell us little about whether the poorest are being left behind—whether the consumption floor is rising above the biological minimum. To address this deficiency, the paper identifies the expected value of the floor as a weighted mean of observed consumptions for the poorest stratum. Under the identifying assumptions and using data for the developing world over 1981–2011, the estimated floor is about half the $1.25 a day poverty line. Economic growth and social policies have delivered only modest progress in raising the floor, despite overall growth and progress in reducing the number living near the floor.

This is a preview of subscription content, log in to check access.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11


  1. 1.

    The press release was for an IFPRI report Ahmed et al. (2007).

  2. 2.

    Quoted by James (2015).

  3. 3.

    The first expression is attributed to John F. Kennedy, the middle claim is the title of an influential paper by Dollar and Kraay (2002), reiterated by Dollar et al. (2013), while the last expression is due to Radelet (2015).

  4. 4.

    The most commonly used axioms are: (i) focus: the measure of poverty should be unaffected by any changes in the incomes (or consumptions) of those who are not deemed to be poor ; (ii) monotonicity: holding all else constant, the measure of poverty must rise if a poor person experiences a drop in her income; (iii) subgroup monotonicity: poverty rises when any sub-group becomes poorer; (iv) scale invariance: the measure is unchanged when all incomes and the poverty line increase by the same proportion; (v) the transfer principle: the measure of poverty falls whenever a given sum of money is transferred from a poor person to someone even poorer. An influential early contribution to the axiomatic foundations was made by Sen (1976), although Sen’s measure did not satisfy all of the above axioms. Other axioms have also been proposed; for a fuller listing see Foster et al. (2013).

  5. 5.

    For further discussion of poverty lines in theory and practice see Ravallion (2012).

  6. 6.

    This reflects well known limitations of this measure, which fails both the monotonicity and transfer axioms. The income share of the poorest x % has been used as a measure of inequality, but it is also known to have deficiencies when judged by the standard axioms of inequality measurement; for further discussion see Fields (2001, Chap. 2).

  7. 7.

    The limit of the Foster et al. (1984) measure as their inequality aversion parameter goes to infinity is the lowest value level in the data. This is only the floor, as measured here, if one is certain that the lowest observed value is the lower bound to permanent consumption. It is argued below that this is questionable.

  8. 8.

    See, for example, Freiman’s (2012) comments on Rawls’s difference principle.

  9. 9.

    See, for example, the discussion in Fleurbaey and Maniquet (2011, Chap. 12).

  10. 10.

    While popularity need not guide ethical judgments, it is at least notable in the context of understanding debates about distributive justice that there is experimental evidence indicating that a non-negligible number of people make distributional judgments consistently with a Rawlsian “maximin” criterion (Michelbach et al. 2003).

  11. 11.

    The veil of ignorance was a thought device to assure that morally irrelevant—inherited or acquired—advantages in the real world did not color judgments about distributive justice.

  12. 12.

    There are also well-known efficiency arguments, notably in non-competitive labor markets. The first minimum wage law was introduced by New Zealand in 1894.

  13. 13.

    An early example was the Speenhamland System of 1795 introduced by the justices of Berkshire, which guaranteed local working-class residents a basic income indexed to the price of bread; for further discussion of this and subsequent policies see Ravallion (2016, Chap. 10).

  14. 14.

    On the arguments for a social security floor see Cichon and Hagemejer (2007).

  15. 15.

    A good working definition is: “Social safety nets are non-contributory transfers designed to provide regular and predictable support to targeted poor and vulnerable people.” (World Bank 2014, p. xii.) Also see Barrientos (2013).

  16. 16.

    The Di Bao program makes transfers to bring urban residents up to locally determined “Di Bao lines” (see, for example, Ravallion 2014b). The Rural Employment Guarantee Scheme in India aims to guarantee up to 100 days of work per household per year doing unskilled manual labor at stipulated minimum wage rates; see Dutta et al. (2014). The latter program can be interpreted as an attempt to enforce the minimum wage rate in an informal economy.

  17. 17.

    See Blaug’s (1962) discussion of the classical model of wage determination.

  18. 18.

    See Dasgupta (1993, Chap. 13). Blackorby and Donaldson (1984) proposed that social welfare increases with a larger population if and only if the extra people have a level of consumption above a critical minimum. This can be interpreted as an ethical floor, unlike the consumption floor, which is a positive concept.

  19. 19.

    See, for example, Azariadis (1996), Ben-David (1998) and Kraay and Raddatz 2007.

  20. 20.

    An example is found in Lopez and Servén (2009), who add a subsistence consumption parameter to the type of model discussed in Philippe et al. (1999).

  21. 21.

    Examples include Mirrlees (1975), Stiglitz (1976), Dasgupta and Ray (1986), Lipton (1988) and Banerjee and Newman (1994).

  22. 22.

    See, for example, Galor and Zeira (1993). A review of alternative model scan be found in Azariadis (2006).

  23. 23.

    The sources and estimation methods are described in greater detail in Chen and Ravallion (2010).

  24. 24.

    The version of the data set used here is for November 2014.

  25. 25.

    Notice that the use of a group average here is motivated by the data problem, namely that we do not observe the lowest level of permanent consumption. The aim is to estimate that level with the (imperfect) data on hand.

  26. 26.

    For example, for a uniform density \(f_j (y_i )\) in (3), \(\phi (y_i )\) is automatically a decreasing convex function.

  27. 27.

    I tested the approximation using micro data for Bihar, India in 2010. The actual \(P_3 \) was 0.022; the approximation using (12) gave 0.023.

  28. 28.

    Ravallion (2016, Chap. 7) reviews the evidence on the economic gradients in demographic variables.

  29. 29.

    This is the un-weighted mean over time. The inter-temporal variance is so low that it is unlikely that population weighting would make any detectable difference.

  30. 30.

    Ravallion et al. (2009) used Hansen’s (2000) estimator for a piece-wise linear (“threshold”) model in estimating the relationship between national poverty lines and private consumption per person.

  31. 31.

    One might assume instead that nobody above median could be living at the floor. Then \(y^{*}=\$2.00\) per day (for 2005), giving a time-mean of the floor of $0.90. However, the median seems an implausibly high value for \(y^{*}\).

  32. 32.

    This is probably an overestimate given that PovcalNet uses grouped data for many countries, which require curve fitting; the software uses parameterized Lorenz curves fitted to the grouped data. These will give non-zero estimates to very low levels, even when the micro data do not indicate any observations.

  33. 33.

    The trend coefficient is very close to zero (a coefficient of 0.0002, with a standard error of 0.0009).

  34. 34.

    Observations confined to cases with a positive count of those living below $1.25 a day.

  35. 35.

    There is extra noise in the first few survey rounds when a number of survey design and implementation issues were still being resolved. So some averaging is needed.

  36. 36.

    That line was chosen to synchronize with the poverty rate for 1990 implied by the Chen and Ravallion (2010) “$1 a day” line. For further discussion see Ravallion (2014a).

  37. 37.

    This is true of Australia, Austria, Belgium, Canada, Czechoslovakia, Germany, Japan, Hungary, Korea, Luxembourg, New Zealand, the Scandinavian countries, Switzerland and Taiwan. Ravallion (2014a) gives the results. Exceptions are the United Kingdom and the United States, with 2.5 and 3.0 % still living in extreme poverty by 1992. While not strictly comparable, the estimates in Ravallion and Chen (2013) suggest that at some time after 1990 the number of people living below $1.25 a day in the high-income countries went to zero.

  38. 38.

    In the country groupings used by Bourguignon and Morrisson the consumption floor in 1850 was 45 % of the poverty line in Australia–Canada–New Zealand, 51 % in Austria–Czechoslovakia–Hungary, 54 % in Belgium–Luxembourg–Switzerland, 55 % in Germany, 41 % in Japan, 42 % in Korea–Taiwan and 49 % in Scandinavia.

  39. 39.

    White standard errors in parentheses. Using the smaller data set from Ravallion et al. (2009) the corresponding estimate of the floor was $0.647 (s.e. = 0.288; n = 73).

  40. 40.

    The preferred functional form in Ravallion et al. (2009) is not feasible since it uses a log transformation of \(\bar{{y}}_i\) for which \(E(z_i \left| {\bar{{y}}_i =0)} \right. \) is undefined.

  41. 41.

    The CDF is truncated above $20 a day to give greater detail at the lower end; however, there is first-order dominance over the entire range.

  42. 42.

    On the implications of first-order dominance in this context see Atkinson (1987).

  43. 43.

    The empirical quantile function is used for 1981. For the purpose of creating the graph, the quantile function for 2011 was based on a 10th degree polynomial, which fitted extremely well \((\hbox {R}^{2}=0.998)\), although the top 2 % were trimmed as these are considered less reliable.

  44. 44.

    The use of the 95 % confidence interval is essentially arbitrary. I also give results for other lines.

  45. 45.

    For the $0.77 line the annual trend is \(-\)0.69 % (s.e. = 0.05).

  46. 46.

    On the history of social protection policies in the U.K. and U.S. see Mencher (1967).

  47. 47.

    See World Bank (2014). For South Asia the overall coverage rate is 25 %, for MENA it is 28 %, for East Asia it is 48 % while for EECA it is 50 %.

  48. 48.

    The regression coefficient of the log of coverage rate for the poor on the log of GDP per capita is 0.91 with a standard error is 0.13. The corresponding elasticity for the population as a whole is 0.80 (s.e. = 0.11). If one controls for the overall coverage rate of the population there is no longer any statistically significant effect of GDP on the coverage rate of the poorest quintile.

  49. 49.

    Regressing the log of the ratio of coverage rate for the poor to the overall coverage rate on the log of GDP per capita gives a regression coefficient of 0.16, with a standard error of 0.04.


  1. Ahmed, A., Hill, V., Ruth, S., Lisa, C., Wiesmann, D. M., & Frankenberger, T. (2007). The world’s most deprived. Washington, DC: International Food Policy Research Institute.

    Google Scholar 

  2. Atkinson, A. B. (1987). On the measurement of poverty. Econometrica, 55, 749–764.

    Article  Google Scholar 

  3. Azariadis, C. (1996). The economics of poverty traps. Part one: Complete markets. Journal of Economic Growth, 1, 449–486.

    Article  Google Scholar 

  4. Azariadis, C. (2006). The theory of poverty traps: What have we learned? In S. Bowles, S. Durlauf, & K. Hoff (Eds.), Poverty traps. Princeton: Princeton University Press.

    Google Scholar 

  5. Banerjee, A., & Newman, A. (1994). Poverty. Incentives and development. American Economic Review, 84(2), 211–215.

    Google Scholar 

  6. Barrientos, A. (2013). Social assistance in developing countries. Cambridge: Cambridge University Press.

    Google Scholar 

  7. Ben-David, D. (1998). Convergence clubs and subsistence economies. Journal of Development Economics, 55, 153–159.

    Article  Google Scholar 

  8. Blackorby, C., & Donaldson, D. (1984). Social criteria for evaluating population change. Journal of Public Economics, 25, 13–33.

    Article  Google Scholar 

  9. Blaug, M. (1962). Economic theory in retrospect. London: Heinemann Books.

    Google Scholar 

  10. Bourguignon, F., & Morrisson, C. (2002). Inequality among world citizens: 1820–1992. American Economic Review, 92(4), 727–744.

    Article  Google Scholar 

  11. Bowley, A. L. (1915). The nature and purpose of the measurement of social phenomena. London: P.S. King and Sons.

    Google Scholar 

  12. Chen, S., & Ravallion, M. (2010). The developing world is poorer than we thought, but no less successful in the fight against poverty. Quarterly Journal of Economics, 125(4), 1577–1625.

    Article  Google Scholar 

  13. Cichon, M., & Hagemejer, K. (2007). Changing the development policy paradigm: Investing in a social security floor for all. International Social Security Review, 60, 169–196.

    Article  Google Scholar 

  14. Dasgupta, P. (1993). An inquiry into well-being and destitution. Oxford: Oxford University Press.

    Google Scholar 

  15. Dasgupta, P., & Ray, D. (1986). Inequality as a determinant of malnutrition and unemployment. Economic Journal, 96, 1011–1034.

    Article  Google Scholar 

  16. Dollar, D., & Kraay, A. (2002). Growth is good for the poor. Journal of Economic Growth, 7(3), 195–225.

    Article  Google Scholar 

  17. Dollar, D., Kleineberg, T., & Kraay, A. (2013). Growth still is good for the poor. Policy Research Working Paper 6568, World Bank.

  18. Dutta, P., Murgai, R., Ravallion, M., & van de Walle, D. (2014). Right-to-work? Assessing India’s employment guarantee scheme in Bihar. Washington, DC: World Bank.

    Google Scholar 

  19. Fields, G. (2001). Distribution and development. New York: Russell Sage Foundation.

    Google Scholar 

  20. Fleurbaey, M., & Maniquet, F. (2011). A theory of fairness and social welfare. Cambridge: Cambridge University Press.

    Google Scholar 

  21. Freiman, C. (2012). Why poverty matters most: Towards a humanitarian theory of social justice. Utilitas, 24(1), 26–40.

    Article  Google Scholar 

  22. Friedman, M. (1957). A theory of the consumption function. Princeton, NJ: Princeton University Press.

    Google Scholar 

  23. Foster, J., Greer, J., & Thorbecke, E. (1984). A class of decomposable poverty measures. Econometrica, 52, 761–765.

    Article  Google Scholar 

  24. Foster, J., Seth, S., Lokshin, M., & Sajaia, Z. (2013). A unified approach to measuring poverty and inequality. Washington, DC: World Bank.

    Google Scholar 

  25. Galor, O., & Zeira, J. (1993). Income distribution and macroeconomics. Review of Economic Studies, 60(1), 35–52.

    Article  Google Scholar 

  26. Hammond, P. (1976). Equity, arrow’s conditions and Rawls’ difference principle. Econometrica, 44, 793–804.

    Article  Google Scholar 

  27. Hansen, B. E. (2000). Sample splitting and threshold estimation. Econometrica, 68, 575–603.

    Article  Google Scholar 

  28. Harsanyi, J. (1975). Can the maximin principle serve as a basis for morality? A critique of John Rawls’s theory. American Political Science Review, 69(2), 594–606.

    Article  Google Scholar 

  29. International Labor Organization. (2012). “R202:Social Protection Floors Recommendation,” Recommendation concerning National Floors of Social Protection Adoption, Geneva, 101st ILC session (14 June 2012).

  30. James, A. (2015). “World’s poorest “left behind”, says Vatican,” Christian Radio 11 February

  31. Jolliffe, D., & Prydz, E. (2015). “Global poverty goals and prices. How purchasing power parity matters,” Policy research working paper 7256, World Bank.

  32. Kraay, A., & Raddatz, C. (2007). Poverty traps, aid and growth. Journal of Development Economics, 82(2), 315–347.

    Article  Google Scholar 

  33. Lewis, A. (1954). Economic development with unlimited supplies of labor. Manchester School of Economic and Social Studies, 22, 139–191.

    Article  Google Scholar 

  34. Lindgren, M. (2015). “The Elusive quest for the subsistence line. How much does the cost of survival vary between populations?” Comparative institutional analysis working paper 2015:1, Lund University, Sweden.

  35. Lipton, M. (1988). “The poor and the ultra-poor: Some interim findings,” Discussion Paper 25. Washington, DC: World Bank.

  36. Lopez, H., Servén, L. (2009), “Too poor to grow,” Policy research working paper 5012, World Bank.

  37. Maddison, A. (1995). Monitoring the world economy. Paris: OECD.

  38. Malthus, T. R. (1806). An essay on the principle of population (1890th ed.). London: Ward, Lock and Co.

    Google Scholar 

  39. Mencher, S. (1967). Poor law to poverty program: Economic security policy in Britain and the United States. Pittsburgh: University of Pittsburgh Press.

    Google Scholar 

  40. Mirrlees, J. (1975). A pure theory of underdeveloped economies. In L. G. Reynolds (Ed.), Agriculture in development theory. New Haven: Yale University Press.

    Google Scholar 

  41. Michelbach, P. A., Scott, J. T., Matland, R. E., & Bornstein, B. H. (2003). Doing Rawls justice: An experimental study of income distribution norms. American Journal of Political Science, 47(3), 523–539.

    Article  Google Scholar 

  42. Radelet, S. (2015). The great surge: The unprecedented economic and political transformation of developing countries around the world. New York: Simon & Schuster.

    Google Scholar 

  43. Philippe, A., Eve, C., & Cecilia, G.-P. (1999). Inequality and economic growth: The perspectives of the new growth theories. Journal of Economic Literature, 37(4), 1615–1660.

    Article  Google Scholar 

  44. Ravallion, M. (2012). Why don’t we see poverty convergence? American Economic Review, 102(1), 504–523.

    Article  Google Scholar 

  45. Ravallion, M. (2014a). “Poverty in the rich world when it was not nearly so rich,” Blog post, Center for Global Development, Washington, DC.

  46. Ravallion, M. (2014b). An emerging new form of social protection in 21st century China. In S. Fan, R. Kanbur, S. Wei, & X. Zhang (Eds.), Oxford companion to the economics of China. Oxford: Oxford University Press.

    Google Scholar 

  47. Ravallion, M. (2014c). Income inequality in the developing world. Science, 344, 851–855.

    Article  Google Scholar 

  48. Ravallion, M. (2016). The economics of poverty, history, measurement and policy. New York: Oxford University Press.

    Google Scholar 

  49. Ravallion, M., & Chen, S. (2003). Measuring pro-poor growth. Economics Letters, 78(1), 93–99.

    Article  Google Scholar 

  50. Ravallion, M., & Chen, S. (2013). A proposal for truly global poverty measures. Global Policy, 4(3), 258–265.

    Google Scholar 

  51. Ravallion, M., Chen, S., & Sangraula, P. (2009). Dollar a day revisited. World Bank Economic Review, 23(2), 163–184.

    Article  Google Scholar 

  52. Rawls, J. (1971). A theory of justice. Cambridge, MA: Harvard University Press.

    Google Scholar 

  53. Roemer, J. (1996). Theories of distributive justice. Cambridge: Harvard University Press.

    Google Scholar 

  54. Roemer, J. (2014). Economic development as opportunity equalization. World Bank Economic Review, 28(2), 189–209.

    Article  Google Scholar 

  55. Sen, A. (1976). Poverty: An ordinal approach to measurement. Econometrica, 46, 437–446.

    Google Scholar 

  56. Stiglitz, J. E. (1976). The efficiency wage hypothesis, surplus labour and the distribution of income in L.D.Cs. Oxford Economic Papers, 28, 185–207.

    Google Scholar 

  57. United Nations. (2011). The Millennium Development Goals Report. New York: United Nations Secretariat.

  58. World Bank. (2014). The State of Social Safety Nets 2014. Washington, DC: World Bank.

Download references

Author information



Corresponding author

Correspondence to Martin Ravallion.

Additional information

For helpful discussions on the topic of this paper and comments on the paper the author is grateful to Francois Bourguignon, Mary Ann Bronson, Cait Brown, Shaohua Chen, Denis Cogneau, James Foster, Garance Genicot, Peter Lanjouw, Nkunde Mwase, Henry Richardson, Dominique van de Walle and seminar participants at the International Monetary Fund, the Paris School of Economics, the International Labour Organization, the University of Antwerp, the Human Development and Capabilities Association Conference 2015, Georgetown University, and the 30th annual conference of WIDER, Helsinki, 2015. The author also thanks the journal’s editor and five referees for their comments. One of the referees was especially helpful in developing a stronger rationale for the proposed estimator in Sect. 4.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary material 1 (xlsx 10 KB)

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Ravallion, M. Are the world’s poorest being left behind?. J Econ Growth 21, 139–164 (2016).

Download citation


  • Poverty
  • Consumption floor
  • Rawls
  • Growth
  • Safety-nets


  • I32
  • I38
  • O15