Clarification needed: Measuring Extra Costs of Disability

As we mentioned in our October 2017 blog entry, a growing body of qualitative and quantitative research confirms what is well known to people with disabilities – living with a disability entails extra costs. Not accounting for these costs leads the inadequacy of social protection programs or other programs designed to promote inclusion, an issue growingly acknowledged in recent years, especially by social protection programs and professionals.

However, we realized during our training and advising work that there is a risk of confusion in understanding issues around extra costs and related measurement methodologies. The key issue being: Are we measuring the current economic impact of disability on individuals and their families, or are we estimating the cost and expenditures that would be required for them to participate on equal basis with other members of the community. This is a critical difference that we would like to reiterate and explore a bit further in this blog.

Different type of costs: Some disability related cost are indirect costs – such as foregone income of people with disabilities who face barriers to employment or foregone income of family members who have added family responsibilities. Others are direct costs, that is expenditures on things like medical care, assistive devices, personal assistance, transportation, etc. This blog focuses on the how to estimate those direct costs.

Different measurement methods exist for estimating the direct costs of disability. Two methods seek to estimates the current disability related cost faced by individuals and their families.

  • Good and services (GS): One way is to ask people with disabilities to report on their disability related  expenditures as well as resource allocation they made due to disability. Getting good, representative data this way is often expensive.  It involves long interviews and/or focus groups with skilled interviewers. Also, it could be that people need goods and/or services that aren’t available, so the expenditures they report may be low not because they don’t have needs but because there is no way to meet them.
  • Standard of living methods (SOL): An much cheaper alternative is  the Standard of Living (SOL) approach which has been the basis of most of the rearsch conducted on the subject in LMICs. While it has the advantage of being much easier to implement – and while it can be very useful – it is important to understand its limitations and how it should be used. In the SOL, the assumption is that families with and without disabilities who have the same level of income, same number of people living in the household, are living in the same geographic are, etc. should have the same standard of living. If they don’t, then the gap in the SOL must be due to the disability related extra expenses as well as resource allocation choices mady by households with member(s) with disability . This gap in the standard of living is measured by the difference in assets. For a technical explanation of how this is done you can see the article listed at the end of this blog.  The key thing, though, is that this method basically seeks to estimate the extra expenditures that households with disabled members currently make.

Often outcome of SOL method analysis is referred to as the extra costs of disability, without much explanation as to what exactly is being measured. This may be misleading (we have done it ourselves and realized the confusion it can create!). As the article I reference explains, the SOL methods seeks to measure current disability related expenditures made by households. In reality these are most likely to be much less than the disability related expenditures required for equal participation, especially for households in the lower part of the income distribution. More expenditures may be needed for goods and services that are not available today or that are available but not affordable, or, maybe families choose not to spend their limited resources in a manner to support inclusion of their family member(s) with disabilities.  In that sense, the SOL approach (as the good and service method) should be considered as method that estimates a lower bound on what costs are needed to ensure full inclusion and should not be used alone to setup a benchmark for social protection or other programs.

A risk exists that the SOL method could be used alone to establish a benchmark without further analysis of the gaps in spending needed to ensure full participation, because SOL is a straightforward, inexpensive and frequently referred to as measuring extra “costs” as opposed to extra “expenditures currently being made.” There is a difference between what is being spent and what should be spent to ensure inclusion.

To estimate the disability related expenditures required for equal participation and inclusion, a different method is needed: the good and services required (GSR). The GSR asks people through surveys and/or focus group what they currently use and what additional expenditures, if any, they would need to participate equally (employment, family and community life…). These expenditures are   priced out in their given context (not only nationally but taking into regional differences). Getting good quality, representative data in this fashion is also resource intensive. It proposes certain methodological challenges, as well.  People may not know what types of goods or services are potentially available. Prices may differ dramatically by where people live. However, it is a necessary investment in order to build policy responses that truly contribute to support participation and inclusion.

So, the SOL measure is still important and useful, but best used in conjunction with the other methods. The SOL is inexpensive and easy to use once good quality questions identifying people with disabilities – preferably the Washington Group questions – are included on standard household income and expenditure surveys. These surveys are routinely done by most countries, so those estimates can be produced every time those surveys are fielded (typically every one to three years). Because they are nationally representative and typically involve thousands if not tens of thousands of households, results can be broken down by type of disability, region, age, gender, or other characteristics to see how extra expenditures vary by people’s situations.

Combined with more detailed studies are done on actual expenditures as reported by households with disabled members, the accuracy and variability of the SOL estimates could be better understood. And then if studies were also done on determining what would actually be needed to equalize standards of living (again, on a less often basis) the expenses needed could be compared to those made to see (a) are the extra costs of living with a disability being met, and (b) does the adequacy of those expenditures vary by type of disability, region, age, gender, etc. These more detailed studies could be done on a much less frequent basis because the relationship between SOL estimates and reported expenditures probably does not change rapidly.

By using different methodologies together – and appropriately – we can combine the lower resource needs and large-scale data collection power of the SOL approach with the higher resource needs but more in-depth and detailed approach of the other methodologies to leverage their respective advantages to get an accurate, comprehensive picture of the extra costs of disability.