Question

A) Explain the relationship between economic growth and income inequality. Based on your answer, explain the phenomenon described by development economists as the Kuznets Inverted U curve.

B) What evidence supported the Kuznets inverted U curve, albeit mostly speculative? [Kuznets readily admitted that the basis for the relationship was “perhaps 5 percent empirical and 95 percent speculation”].

Also, what was W. Arthur Lewis’ model’s evidence and assumption in support of the Inverted U curve?

C) What evidence has been advanced to reject the Kuznets curve in recent times?

Answer 1

As there are many multiple parts for this answer. I will be attaching few pdfs for your reference. There have been many reports and studies for this to understand. Please along with the following brief overview, refer to the attachments and links for more clarity.

- To understand the overview : https://econthoughtdotorg.wordpress.com/2014/10/12/aer-top-20-kuznets-on-inequality-1955/

- For more clarity and detailed analysis : http://www.acarindex.com/dosyalar/makale/acarindex-1423903663.pdf

- https://www.thoughtco.com/kuznets-curve-in-economics-1146122

Introduction

Although economic growth is a desired phenomenon in most countries in the world, the pattern of income distribution is also of importance as high levels of inequality and poverty is detrimental to further growth and development. Economists in the past and up till date have sought to study the relationship between economic growth and income distribution. The first recognised attempt at analysing the relationship between income inequality and income per capita was made by Simon Kuznets (1955).

The Kuznets hypothesis argues that rising levels of inequality are associated with early stages of development and then subsequently falls in the later stages of development i.e. the shift from agriculture to industrialisation. Kuznets’ findings were based on historical data for three developed countries, which were the US, Germany and England, for the first half of the nineteenth century and the assumption that per capita income in the non-agricultural sector is always higher than that of the agricultural sector and that inequality in income distribution within the agricultural sector can only be equal or less than the non-agricultural sector. According to him, based on his assumptions, inequality ought to keep on growing wider with development but his empirical results suggested otherwise (a rise and then a subsequent fall in inequality with increasing per capita income). He attributed this anomaly to an increase in incomes and political power and influence of lower groups in the non-agricultural sector. The graph when plotted of this relationship between income inequality and income per capita has the shape of an inverted U- curve. Moran (2005) describes the Kuznets curve hypothesis as the most influential publication on inequality and development. Several theories have been put forward to explain the inverted U-curve hypothesis. According to Williamson (1991) and Barro (2000), technological change is responsible for the inverted U- curve shape between inequality and development. They argue that income inequality increases when there is a new change in technology and then decreases when this technology becomes well-known. On the other hand, Aghion and Bolton (1997) say that credit market imperfections is responsible for the U- shaped curve. According to their theory, the rich get richer in early stages of development as they are able to easily access credit form the credit market whilst the poor get poorer but in the more advanced stages of development, the interest rate falls as a result of wealth accumulation by the rich which allows the poor to get richer thereby reducing inequality levels.

THE KUZNETS INVERTED U HYPOTHESIS EMPIRICAL EVIDENCE

In the past and in recent times, quite a number of empirical researches have been carried out on the validity of the Kuznets U-curve hypothesis but the results obtained remain inconclusive. Oksana and Jakub (2014), using a sample of 145 countries for the period of 1979 – 2009 observed the inverted U-curve in countries with low amount of social contribution and a flat U-curve with countries with high amounts of social contribution. Ahluwalia (1976) in his research involving 60 countries found support for the Kuznets hypothesis. In contrast, using the same sample and data as Ahluwalia, Anand and Kanbur (1973) did not observe the Kuznets inverted U-shaped curve. Thornton (2001) from panel data analysis using data on Gini coefficients, income quintiles and real GDP per capita on 96 countries over the post-war period found evidence of the Kuznets hypothesis. Hieroms (2010) findings from panel data analysis for Latin America countries from 1975- 2008 suggest that the Kuznets hypothesis is not observed in the region. Deininger and Squire (1998) carried out cross-country analysis and examination of country specific time series. Their results did not support the Kuznets inverted U-curve in both analysis. However, Milanovic (2000) found that Kuznets hypothesis was observed for 80 countries during the 1980s. Likewise, Bulir (2001) also observed the Kuznets hypothesis from analysis of cross-sectional data for 75 countries. Banerjee & Duflo (2003) using non-parametric methods suggest that the growth rate is an inverted U-shaped function of net changes in inequality. Interestingly, Barro (2000) found there to be no evidence of a relationship between inequality and economic growth amongst countries in the world but on grouping these countries into rich and poor and he observed a negative relationship between inequality and growth in rich countries and a positive relationship in poor countries. Fields (1989) also suggests that there is no evidence supporting an increase in inequality with economic development.

Policy implications of the Kuznets Hypothesis

Kuznets’ U-curve hypothesis suggests that there is a trade-off between economic development and growth. In other words, a country experiencing higher levels of GDP per capita should expect rising inequality until a certain threshold is passed and then inequality starts to decline. This school of thought was almost a ‘stylised fact’ and was further buttressed by the findings of Ahluwalia (1976). However, opponents of the Kuznets’ hypothesis argue that Kuznets use of cross-country data instead of panel data and within-country time-series does not fully eliminate country fixed effects and might create the illusion of the Kuznets’ curve (Hieroms, 2010). Hence, there really may not exist a trade- off between both phenomena i.e. a country can enjoy both increasing economic development and reducing level of income inequality. Bourguignon (2004) argues that country specificity in the way economic growth affects distribution plays a significant role, and as a result no generalization as that of the Kuznets hypothesis could be possible. In the past, the Kuznets hypothesis was adopted by the World Bank in making projections about inequality and world poverty (Anand et al., 1993). However, Deininger and Squire (1998) and Anand and Kanbur (1973) results point to the contrary, i.e. higher levels of development are not always associated with increasing inequality. Furthermore, Campano and Salvatore (1988) found evidence of the Kuznets’ curve for the lowest quintile of their sample, suggesting, that the U- hypothesis differs amongst groups in the society. Ram (1988) also suggests that the observation of the U-hypothesis depends on the region. Also implied from Kuznets (1955) is that developing countries will be characterised by increasing inequality and developed countries by decreasing inequality but Gallup (2012) using panel data observed an ‘anti-Kuznets curve’, i.e. developing countries were experiencing decreasing levels of inequality and the developed countries increasing levels of inequality. This finding gives credence to Hieroms (2010) stance on the method of analysis bias. Such conflicting empirical results suggest that the Kuznets hypothesis is not infallible (It does not hold for all countries) and as such should be taken with a pinch of salt in policy formulation in countries. Kuznets, himself, concluded his 1955 publication by saying that 5% of it was based on facts and the other 95% on speculation.

Conclusion

Two major schools of thought exist on the relationship between inequality and development, the classical view is that inequality is beneficial to long term growth because the incomes of the higher groups in society will be channelled towards investment as they have a higher propensity to save compared to the other classes in society. On the contrary, the second view is that inequality impedes development since it results in a loss of human capital as a result of the imperfections in the credit market which hinders access to loans by the lower groups in society. However, there seems to exist a consensus in the literature that high income inequality in a society is detrimental to long term economic development because it results in political instability, increased crime etc.

From the findings from more recent scholars who made use of various estimation techniques, the support for this hypothesis is mixed and results inconclusive. Perhaps, changes in GDP per capita cannot fully explain changes in income inequality. Therefore, policymakers in economies enjoying economic growth should take into account that deliberate action possibly through government action is required to ensure more equitable resource distribution and thus policies focused on income redistribution and welfare improvement should be pursued especially if long term development is desired. Simply accommodating increasing levels of inequality whilst waiting for the Kuznets threshold is risky as even this threshold for individual countries is unknown. This essay suggests that the observation of the Kuznets hypothesis may differ between regions and groups in the society and as such may be as a result of certain peculiarities in such regions. Probably other factors not yet mentioned in the literature such as corruption levels, type of political systems etc. determine the observation of Kuznets curve in a country. Further empirical research which takes into account such regional peculiarities instead of wide cross-sectional investigations is advocated for.