In a guest post, Daniel Goldberg argues in favour of ethically-charged ways of talking about inequality.
The terms we use to describe inequality are important. They’re important because there are different perspectives about (1) what terms should be used for (2) which referents (3) in which circumstances, and these reveal what is actually at stake. Thus, whether the term “disparities” is used in lieu of “inequalities” or “inequities” actually matters a great deal, at least in part because there is an inescapable ethical valence to the term “inequities” that is absent from the more neutral term “disparities.” The question is whether the ethical valence is something that ought to be included in public health policy discourse, or whether, given the difficulty in measuring it, it should be avoided where possible.
Regarding these points, Barbara Starfield, University Distinguished Professor in the Department of Health Policy & Management at Johns Hopkins University, has a fascinating new editorial out in the International Journal for Equity & Health entitled The Hidden Inequity in Health Care. Professor Starfield’s work has been instrumental both to my own thought and, I think it is fair to say, to larger matters of inequities in health and in health care, especially primary health care. Nevertheless, there was one line in the editorial I’d like to pick out and examine, not simply because I respectfully disagree with it, but also because it sheds some light on some larger issues on the role of ethical theory in thinking about health inequities. Professor Starfield writes that:
“Some people use the term ‘unfairness’ to define inequity, but unfairness is not measurable and therefore not a useful term for policy or evaluation” (p. 3).
Although the experimental philosophers may chafe, I’m going to assume for the sake of argument that the premise of this argument is correct, that unfairness is not measurable. Perhaps to sharpen the premise a bit, I might want to assert that Professor Starfield means to use “quantifiability” as the criterion for measurability, since of course we can obtain all sorts of rich empirical data regarding unfairness and health via qualitative methods. But I am going to assume that data cannot be captured in robust quantifiable form.
It is the conclusion of the argument here that I disagree with, but the nature of my disagreement suggests some important points (or at least they are important to me!). I disagree with the idea that if a particular phenomenon is not quantifiable it is not useful in policy or evaluation. Even putting aside the issue of whether utility is the right criterion here, I do not agree that only facts that can be mobilized and calibrated quantitatively are relevant for health policy analysis and evaluation.
Accordingly, I have written elsewhere on what I have termed “the unbearable oughtness of health policy.” Especially when thinking about health and inequalities, even the best descriptive, quantitative data, however important in undeniably is, does not by itself indicate what a just social order ought to do with regards to improving health and compressing inequalities. In my view, reading the normative valence ought of health policy analysis is therefore not an advisable course; the entire question is ‘what kind of people do we wish to be’ – yes, I like virtue ethics – and insofar as we practice our collective answer to that question through public policy (in)action, attempting to excise the ethical valence from the concept of health equity is not a good idea. (I also think it is quite impossible, but that is a post for another day).
I found Professor Starfield’s point here particularly fascinating given that the terminological debate – disparities, inequalities, inequities – in part revolves around the significance of the normative aspect. That is, Sir Marmot has gone on record (video link) indicating he eschews the term “disparities” and prefers “inequities” precisely because the latter carries ethical content; it is not just brute social and health differences among and between populations that matter, but differences that are unfair or unjust. Of course, specifying criteria for unfairness is no easy matter, as Professor Starfield suggests. But some engaged in larger discourse on inequalities – me among them – see the express inclusion of ethical valence in the term “inequity” as a feature rather than as a flaw to be avoided.
Bibliographical Notes: David Frankford has a nice 1994 paper on related issues, addressing the rise of health services research and the apparent objective of avoiding ethical deliberation on issues of justice and fairness in health policy. The history of quantification, and the ways in which it began to assert an increasingly powerful grip on science and health policy is also significant here (for the classic text, see Theodore Porter, Trust in Numbers).