The dilemmas involved for Governments in saving lives
31 March 2020
By Graham Loomes
As the current COVID-19 crisis highlights, governments have to take many difficult health-related decisions that may have profound social and economic implications for their populations.
In arriving at those decisions, trade-offs are inevitable. These trade-offs are often complex, weighing benefits and costs of different magnitudes, at different times and to different recipients.
How do you balance the resources required for particular interventions against the benefits in terms of the health outcomes obtainable? And how should those benefits be distributed across a population?
Such judgements may be informed by the preferences of the population. Different individuals’ self-interest is liable to be an important consideration, but this is unlikely to be the only factor in weighing trade-offs and making policy choices that reflect people’s preferences for the kind of society in which they live.
Individuals might, for example, prefer to focus resources in a way that maximises total overall benefit – and therefore the efficiency of a particular policy intervention – even if some groups (including themselves, possibly) do not directly benefit from that intervention.
On the other hand, they may believe fairness or equity is an important principle in distributing resources and that some level of benefit should be available to everyone, even if the total benefit is smaller and so overall efficiency is (somewhat) lower.
To explain the dilemma, imagine two countries with the same population. In Country X, the total amount of wealth is higher than in Country Y, but it is very unevenly spread, with a few people being very wealthy and quite a lot of people having very low levels of wealth.
Country Y has less wealth in total, so the average wealth in X is higher than the average wealth in Y (meaning that efficiency is higher in X than in Y), but wealth in Y is more evenly spread so that the difference in wealth between the poorest and the wealthiest people in Y is smaller than between the poorest and the wealthiest people in X (meaning that Y is more equitable than X). If you were choosing whether to live in X or in Y, how would you weigh up the extra efficiency of X against the greater equity of Y?
Our research set out to uncover more about the relative importance of self-interest, efficiency and equity in people’s preferences for health benefits. Danae Arroyos-Calvera, Research Fellow at Birmingham University, Rebecca McDonald, Senior Lecturer in Economics at Birmingham, and I teamed up with Judith Covey, Associate Professor in Psychology at Durham University, to investigate such trade-offs.
In doing so, we also hoped to test whether words matched deeds. People might say they believed that some degree of equity would be worth prioritising even to the detriment of their own pure self-interest, for example, but would they actually make choices that benefited others at their own expense?
How do we decide what is the best policy intervention?
To that end, we investigated whether there was a correlation between the principles espoused by individuals, gauged from their responses to questions about trade-offs, and their preferred allocation of resources as indicated by the choices that they made between different policy scenarios.
We used the scenario of a city where the inhabitants were currently at differing risks of adverse health effects caused by bacteria in the water supply. The city was divided into an East and West zone, with more people liable to be harmed by the water supply in the West zone. Half of the study participants were told that they lived in the East zone, while the other half lived away from the city. This made it possible to compare self-interested preferences with those of an impartial ‘outsider’ just like a policymaker might be.
We defined four policy options, outlined in the table below. The policies differed in terms of which bacteria they targeted (and hence which zone they helped), and in the magnitude of the risk reductions. Policy EO (‘East-only’) reduced risk in the East zone by 10 cases. Policy WO (‘West-only’) reduced risk in the West by 20 cases. Policy BC (‘both constant’) gave both zones the same risk reduction: it reduced risk by eight cases in each zone. Lastly, policy BR (‘both relative’) also offered a risk reduction in both areas: by four cases in the zone with the higher baseline (the West), and by three cases in the East.
The biggest total reduction – on that measure, the most efficient – is WO. But here, all the benefit goes to the people in the West and none of it to those in the East. On the other hand, BC gives the same risk reduction to both zones, so in that sense it is the most equitable, but it leaves the total final risk higher (10 + 20) than the total final risk for WO (18 + 8) so it is less efficient.
However, some might say that BC is not fair in another respect, because it gives the East the same reduction as the West without paying attention to the fact that the people in the West are at a higher risk: the baseline is 28 deaths per 100,000 people in the West as compared with 18.
So BR gives reductions that are more (though not completely) proportional to baseline risk: West gets a 4/100,000 reduction while East gets a smaller 3/100,000 reduction. So if the fairness principle is to give a bigger reduction to the people most at risk, BR does that better than BC. But the total gain is quite a bit smaller: (3 + 4) as opposed to (8 + 8). So people may only go for BR over BC if that particular notion of equity matters a great deal to them and they are willing to sacrifice quite a bit of efficiency.
In fact, we found most people strongly preferred BC over BR, but in the choice between BC and WO they were much more evenly divided and found it harder to state a strong preference either way, with those who didn’t live in either area and were therefore more ‘impartial’ slightly favouring the efficiency of WO on average while those who lived in the East and would themselves benefit more from BC were on average a bit more favourable to BC – although the fact that they didn’t all simply strongly favour BC shows that they gave some weight to benefits for people in the West.
The results revealed individuals' broad preferences in terms of the principles that they favoured regarding the distribution of resources. Self-interest was clearly a major factor. However, the results also show that efficiency and equity are both important, with many participants being prepared to sacrifice efficiency, to some extent, in the interests of equity. In addition, the study demonstrated a correlation between the principles that the participants subscribed to and the trade-off choices that they made.
Our findings suggest that it may be possible to use relatively simple instruments to ascertain people’s preferences, use those preferences to infer distributional trade-offs, and then allow for distributional concerns when formulating policy interventions. In due course, we might aspire to link the preferences expressed by individuals quantitatively to financial amounts for cost-benefit analysis purposes. However, this is a subject for future research.
The findings may also have some relevance for the corporate sector. Take the example of a power company that wishes to promote green energy but can only offer that choice at a premium price. Using the same methodology, focusing on principles and trade-offs, the company may be able to get some insights into the willingness of customers to pay more for a product because they value its benefits to the wider population. The same concept can be applied to many other consumer trade-off situations.
But whether it is a public or private setting, one strong message emerges from the research. When considering difficult trade-offs, it would be wrong to assume that everyone is just out for themselves.
Arroyos-Calvera, D., Covey, J., Loomes, G. and McDonald, R. (2019) "The efficiency-equity trade-off, self-interest, and moral principles in health and safety valuation ", Social Science and Medicine, 238, 112477.
For more articles like this download Core magazine here.