Standard of Living and Disability in Cambodia

Abstract

Little is known about the extra costs faced by households with disabled members in low resource settings and the impact of these costs on living standards. In this paper we estimate the direct cost associated with disability for households in Cambodia. Using the Standard of Living approach, the direct cost associated with having a member with disabilities is estimated to be 19 per cent of monthly household consumption expenditure. Accounting for the direct cost of disability doubles the poverty rate amongst households with disabled members from 18 per cent to 37 per cent, and increases the poverty gap from 3 to 8 per cent. A comparison of the direct cost associated with disability and income support received from government and family sources reveals that only 7 per cent of the costs of disability are met. Our findings suggest that, in the absence of increased coverage of public income support, households with disabled members will continue to experience a lower standard of living compared to households without disability in Cambodia.

Acknowledgements

We thank two anonymous reviewers and participants of the Disability Data Workshop (Phnom Penh) and the Australian Health Economics Society conference (Adelaide) for their comments on earlier drafts of this manuscript. The data and code used in this paper are available from the authors upon request.

Disclosure statement

No potential conflict of interest was reported by the authors.

Supplementary Materials

Supplementary Materials are available for this article which can be accessed via the online version of this journal available at https://doi.org/10.1080/00220388.2018.1528349

Notes

1. Disability may also reduce household income because care-giving responsibilities restrict the type or amount of work that family members can do.

2. The literature suggests that low-income countries may have smaller gaps in economic wellbeing between households with and without disabilities compared to middle-income countries because the development process does not benefit people with disabilities. The evidence using a multidimensional measure of poverty is more robust. For example, using a multidimensional measure of poverty that includes measures of employment and education in addition to consumption and assets, Mitra at el. (2013) find that disability is significantly related to poverty in most of the 15 developing countries examined.

3. Seven of the Sustainable Development Goal targets and 14 of the Sustainable Development Goal indicators explicitly mention persons with disabilities, including reducing the proportion of persons with disabilities living below 50 per cent of the median income.

4. There exists limited information on the coverage and efficacy of social protection programmes for persons with disabilities in low- and middle-income countries. Recent evidence from China, Vietnam, Namibia, Zambia, and South Africa suggests that coverage of disability benefits is low and not commensurate with needs (Banks et al., 2017; Hanass-Hancock & McKenzie, 2017; Kumitz, 2015; Palmer, Groce, Mont, Nguyen, & Mitra, 2015). For example, in Vietnam income support for households with disabled members was estimated at between 13 per cent and 26 per cent of the minimum wage (Palmer et al., 2015); in South Africa the average disability grant received by households with more severely disabled members covered 30 per cent of the earned income gap between households without disabled members (Hanass-Hancock & McKenzie, 2017).

5. This approach has been used to value the costs of non-market goods; for example, life events such as unemployment and divorce, health, air pollution, informal care-giving, and violent crime (for example Blanchflower & Oswald, 2004; Groot & van-Den-Brink, 2006; Johnston, Shields, & Suziedelyte, 2015).

6. Mortality estimates during the Khmer Rouge period vary widely (Dy, 2007; Heuveline, 1998). Heuveline (1998) estimates a median of 2.5 million excess deaths over the period 1970–1979 of which 1.4 million were violent deaths with 1.1 million violent deaths occurring during the Khmer Rouge period.

7. A notable legacy of Cambodia’s prolonged civil conflict as relates to disability is one of the highest concentration of land mines in the world (Cambodian Mine Action Centre, 2014; Merrouche, 2011).

8. This prevalence is based on persons experiencing ‘a lot of difficulty’ or ‘cannot do’ in at least one functional domain (http://www.washingtongroup-disability.com).

9. In practice, this is done by regressing consumption expenditure on an indicator for disability.

10. The CSES was first fielded in 1993, and has been conducted annually since 2007. Disability information has been collected in the survey since 2007. However, it was not until the 2009 wave that the structure of the disability questions aligned with the International Classification of Functioning, Disability and Health, enquiring about difficulties in domains of functioning, and the degrees of difficulty in functioning (Palmer & Harley, 2012; World Health Organization, 2001). For this reason, our analysis uses the 2009–2014 waves of the CSES only.

11. The CSES disability questions are in the spirit of, but not identical to, those developed by the United Nation’s Washington Group on Disability Statistics – Short Set of Questions on Disability – which aim to represent the majority of persons with limitation in basic actions (http://www.washingtongroup-disability.com).

12. The Washington Group Short Set of Questions on Disability asks ‘Do you have a difficulty seeing even when wearing glasses?’, so that those who had difficulty seeing that is corrected by wearing glasses would respond they had no difficulty (http://www.washingtongroup-disability.com).

13. The cap on three functional domains for respondents precludes us from constructing a functional score which better captures the variation of health deprivations as relates to wellbeing (Mitra, 2018).

14. Composite measures of different items (typically household assets) are recommended as they reduce the risk of single item preferences or tastes being systematically related to disability status (Zaidi & Burchardt, 2005). We do not include durable assets that are likely to generate income such as ownership of productive land and cattle or a computer.

15. These results are available by request.

16. Disability costs in absolute terms were calculated as $Y\left[exp\left(-\frac{\beta }{\alpha }D\right)-1\right]$. The poverty lines have recently been updated to account for differences in minimum living standards across geographical regions and were derived from the 2009 CSES consumption data used in this study (Ministry of Planning, 2013).

17. The question asked in the CSES is ‘How much did your household receive from ..[SOURCE].. during the last 12 months?’ where the three relevant sources are: (1) Pensions, social welfare/benefits, provident fund; (2) Transfers (assistance/support) from NGO or other institutions (not credit); and (3) Remittances from other relatives or others.

18. There is no state pension provision in Cambodia except for retired civil servants and war veterans.

Additional information

Funding

This work was supported by The University of Melbourne under the Faculty of Business and Economics Research Grant and the Melbourne Social Equity Institute Interdisciplinary Seed Grant.