Growth is Good for the Poor? Not Necessarily

ABSTRACT

In mainstream economic thinking, it is common to find a simplistic relationship between economic growth and poverty reduction. Two decades ago, a paper titled “Growth is Good for the Poor” purported to have established a strict one-to-one relationship between economic growth at the national level and growth of income among the poor. In this note we explain the methodological, modelling, and analytical shortcomings of the exercise. Using simulations based on random numbers we show that the data did not support the arguments and that, consequently, the policy conclusions were not warranted by the analysis. The lessons are important in the current context when addressing growing poverty in the wake of the COVID-19 pandemic.

KEYWORDS:

• Economic growth, income distribution, poverty, monetary poverty, development, simulation statistical analysis

JEL CODES:

• B41
• D31
• I32
• O11
• O15

Acknowledgements

The authors wish to thank L.P. Rochon and two anonymous referees for their encouragement and valuable comments to an earlier draft of this article.

Disclosure Statement

No potential conflict of interest was reported by the author(s).

Notes

1 In a way reminiscent of the critique of production functions and income distribution models by Shaikh (1980)

2 We have also updated the range for the variation of per capita income and the share of income going to the bottom quintile to be similar to current values. Our results remain the same.

3 Clearly, when carrying out this type of random simulation, one run is not sufficient. Therefore, we ran hundreds of them. We have also varied the ranges of values for each of the three series. The results are always consistent with the pattern described in the text.

4 In other words, this is similar to a case of spurious correlation when two independent random variables are divided by a third random variable (Everitt and Skrondal 2010).

5 The results in Diagram 1 correspond to values for GDP per capita between 270 and 85,000 dollars per year which are more realistic numbers than those used two decades ago. The range for the share of income of the bottom 20 per cent was retained, as it is consistent with the latest data from the UNU WIDER database on income distribution (https://www.wider.unu.edu/project/wiid-%E2%80%93-world-income-inequality-database).

6 If it were true that a 1 percent increase in average income reduces poverty by 1 percent, it only takes projecting backwards in time from current levels of poverty incidence to find a time when everybody (even kings and business magnates) was poor (Delamónica 2016).