THE UNIT OF ANALYSIS IN POVERTY MEASUREMENT: HOUSEHOLD OR FAMILY

THE UNIT OF ANALYSIS IN POVERTY MEASUREMENT: HOUSEHOLD OR FAMILY

Jonathan Bradshaw

One of the debates occurring in the field of poverty measurement is what unit of resources should be used? Should resources be measured at the household[1], or benefit unit/family[2], or the individual level? At present the DWP Households Below Average Income series (and the Eurostat EU-SILC series) measure poverty at the household level – the incomes of everyone in the household are added together, an equivalence scale (the subject of another different debate) is applied and the equivalised disposable household income is compared with a threshold (conventionally 60% of the median) and all those living in a household with equivalent incomes below that threshold are defined as living in poverty. The implicit assumption is that that income is equitably shared within the household. But there is evidence that this assumption is not always correct[3] – parents particularly mothers may sacrifice their living standards for children[4], and male breadwinners may not share their earnings fully with their wives and partners[5]. In households containing more than one benefit unit this becomes even more problematic. Will an elderly parent living with her adult children share her income with them and vice versa? Will an older (independent) child still living with his parents share his wages with the parents and vice versa?

           
This note assesses what difference it would make to the child poverty rate if poverty was measured at a benefit unit level. It uses the Family Resources Survey 2015/16 (the latest available)[6]. The poverty rate is estimated as the percentage of children living in households with total equivalised disposable income below a threshold of 60% median after housing costs. The data is weighted to the UK child population (using GS_NEWCH).

The Table shows the number of children by the number of benefit units in the household - 86% of all children live in single benefit unit households, (88% in couple households and 81% in lone parent households). The Table also shows the child poverty rate for multi benefit unit households, single benefit unit households and all households. The child poverty rate is four percentage points higher in all multi benefit unit households than in single benefit unit households. For couples with children the difference is three percentage points higher but for lone parents the poverty rate it is four percentage points lower in multi-unit households. For pensioner couples the child poverty rate is also higher in single benefit unit households. The single pensioner households have very small numbers of children.

 Distribution of children by family type and benefit unit and child poverty rates

Number of Benefit Units within the household		Family type for the benefit unit										Total
		Pensioner couple		Single male pensioner		Single female pensioner		Couple with children		Single with children
1		33281		5453		7718		9042315		2583380		11672147
2		17181		0		1161		1043484		413654		1475480
3		0		2119		0		179380		155355		336854
4		0		0		0		11154		21713		32867
5		0		0		0		3546		13138		16684
6		0		0		0		0		7100		7100
Total	50462		7572		8879		10279879		3194340		13541132
% of children in households with only one benefit unit		66		(72)		(87)		88		81		86
Child poverty rate %
More than one benefit unit households		13		(0)		(100)		27		44		33
Single benefit unit household		21		(73)		(66)		24		48		29
All		18		(53)		(70)		24		47		30

( ) indicates small numbers.

 Having more than one benefit unit in a household alters the equivalence scale and probably in most cases the household income. In couple households the extra benefit units are not making an additional contribution to income beyond their equivalent needs and dragging the household into poverty. In single parent households the additional benefit units are more than compensating for their additional needs. Overall, the more benefit units in a household the higher the poverty rate (single benefit unit=29%, two benefit units =33%, three benefit units=34%, four benefit units=35%).

If we moved from the household to the benefit unit for estimating child poverty rates the overall child poverty rate would increase but the composition of poor children would change with fewer poor children living in lone parent families and more in couple families.

---

[1] A household in the Family Resources Survey is defined as ‘One person living alone or a group of people (not necessarily related) living at the same address who share cooking facilities and share a living room, dining room or dining area, A household will consist of one or more benefit units/families’.

[2] A family or benefit unit is defined as ‘a single adult or a couple living as married and any dependent children’. https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/599163/households-below-average-income-quality-metholodogy-2015-2016.pdf

[3]Gardiner, K. and Millar, J., (2006) How low-paid employees avoid poverty: An analysis by family type and household structure. Journal of Social Policy, 35 (3), pp. 351-369.

See also http://sticerd.lse.ac.uk/case/_new/research/Intra-household/

[4] Main, G. and Bradshaw, J. (2016) Child poverty in the UK: Measures, prevalence and intra-household sharing,  Critical Social Policy, Vol. 36(1): 1–24.

[5] Pahl, J. (1989) Money and Marriage, Macmillan: Basingstoke.

[6] Actually the HBAI data set.