How to help the poor? Go figure

Jan 23, 2014


By Peter A. Coclanis For The Straits Times


QUESTIONS relating to poverty and rising levels of inequality are popping up with increasing frequency all over the developed world.

Concerns over poverty and inequality have roiled Singapore as well. Indeed, an increasingly lively public debate has arisen over these issues, their causes and concomitants. Few domestic issues are as important or potentially explosive as poverty and inequality. Few are also more difficult to get a handle on. It is extremely hard to reach a consensus on how to define, much less measure, poverty and inequality. It is even more difficult to understand the root causes and the most effective approaches to amelioration.

Whether one refers to minimum income thresholds, basic material needs, or civic inclusion and capabilities goes a long way in determining what proportion of a population will be considered impoverished. Then there is the question of whether to set a hard-and-fast poverty line or employ broader and more elastic concepts such as poverty ranges and bands in an attempt to include populations clustered around (above or below) some level of income or capability.

Even when there is agreement on the proper criteria to employ in calculating poverty levels, accurate measurement - and interpretation - of poverty figures will still depend a lot on demographic factors. Demography may not be destiny but the age structure of the population in question and the population in the denominator can make huge differences in poverty assessments.

In a country such as Singapore, household income figures would likely change significantly if households headed by retirees - currently about 6 per cent of the total number - were included.

If "households" headed by temporary workers were included, the numbers would also change dramatically.

The most common measurement tool of inequality - the Gini coefficient - can be manipulated or misconstrued by the treatment or non-treatment of supplemental benefits, subsidies, tax rebates, and transfer payments as "income". The Gini coefficient ranges from 0 to 1, with 0 denoting perfect equality and 1 perfect inequality. Singapore's Gini coefficient of income inequality is generally considered high and growing, but, even so, it ranges widely from a high of about 0.48 to a low of about 0.41 - a big difference to experts on inequality - depending upon how "income" is defined.

Interpreting inequality, measured via Gini coefficients or otherwise, is no small task either. Here again, the population age structure matters, as do overall income and wealth levels.

Certainly, too much inequality can be a bad thing. But how bad it is differs considerably in countries with high and with low median incomes. Furthermore, it is important to bear in mind that a country can be characterised by high inequality and high rates of economic mobility. This was the case in parts of the United States, for example, at various times in the 19th century.

Given the above, it is not surprising that the causes and consequences of poverty and inequality in the developed world have sparked robust debate as well. Among the causes most commonly cited are technological change and globalisation, which have hit less educated and less skilled segments of the population hard.

Another common argument is that the poor often bear part or all of the blame for their plight because of their poor decisions or behaviour. In the US, many also believe the position of the poor, whatever got them there, has been exacerbated by governmental indifference to and disinvestment in those at the bottom.

Profound problems such as those outlined above cannot be solved in an opinion piece. But a line of research from the US might prove useful. Several important studies have found that certain factors seem to offer some protection against falling into poverty. In a much talked about 2009 study titled Creating An Opportunity Society, Ron Haskins and Isabel Sawhill of the Brookings Institution found that behaviour and "playing by the rules" matters.

Using US census data from 2007, they found that the chances of a household being in poverty were very low when: (1) the head of the family (if not disabled or elderly) completed high school and worked full time; (2) families with children were married (or widowed); and (3) the family head waited until at least the age of 21 before having children.

Using Haskins and Sawhill's sample, comprising over 237,000 individuals, there was "a 98 per cent chance of escaping poverty if the family adheres to all three social norms". Conversely, "76 per cent of those living in families that did not adhere to any of these norms were poor".

In another similar study based on 2012 census data, researchers at the Beverly Lahaye Institute found that only 7.5 per cent of families headed by two married parents lived in poverty, while over a third of families headed by a single mother lived in poverty.

Such studies are not the last word on the subjects they treat, and they probably tell us more about correlations than about causation. That said, it would be interesting if researchers replicated the above approaches in Singapore. The particular criteria employed might be adjusted somewhat to fit Singapore's circumstances - completion of O levels instead of high school graduation, for example - or even changed to include, for instance, membership in a religious body or community organisation. But more information about the role of individual and household behaviour, particularly regarding the degree of adherence to important social norms, may help in policy formulation, whatever the role of technological change and globalisation in the greater scheme of things.

stopinion@sph.com.sg

The writer is the Albert R. Newsome Distinguished Professor of History and Director of the Global Research Institute at the University of North Carolina-Chapel Hill in America. He was Raffles Professor of History for a semester in 2005 at the National University of Singapore.