Article Text

Download PDFPDF

COVID-19: exposing and amplifying inequalities
  1. Michael Marmot,
  2. Jessica Allen
  1. Correspondence to Michael Marmot, Department of Epidemiology and Public Health, UCL, UCL Institute of Health Equity; m.marmot{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Exponential growth is difficult for people to grasp. But that is what has happened to sales of Albert Camus’s The Plague, first published in 1947. According to Jacqueline Rose, it is ‘an upsurge strangely in line with the graphs that daily chart the toll of the sick and the dead’. She reports that, from the start of the COVID-19 pandemic, sales had grown 1000%.1 It may not be worth dwelling on those statistics. More interesting for Rose, and for us, is that a key theme of Camus is that ‘the pestilence is at once blight and revelation. It brings the hidden truth of a corrupt world to the surface’. In the same way, the pandemic of COVID-19 exposes and amplifies inequalities in society. The myth of the pandemic as the great leveller was given air when early cases included elites: a prince, a prime minister, a Premier League football manager and the actor Tom Hanks. It was, and is, most likely that as the pandemic took hold and society responded we would see familiar inequalities, of two sorts: inequalities in COVID-19 and inequalities in the social conditions that lead to inequalities in health more generally.

It was not always thus with epidemics. The plague came to Northern Italy in 1630, killing 35% of the population, including 38% in Bergamo, and an astonishing 59% in Padua. One effect of killing so many people was a temporary slowdown in what had been a steep rise in economic inequality in Italy. In the aftermath of the plague, work was plentiful—so many workers had died—and real wages increased. Property was available at relatively low cost, given how many potential purchasers had also gone, making it easier for lower strata of the population to acquire property. It did not last. By 1650, inequality was again on its relentless rise in Venice, Northern Italy and Italy as a whole.2

Serious as is COVID-19, the worst-case scenario, with no intervention, was perhaps 400 000 deaths in the UK. Terrible as is premature death coming to 0.6% of the population, it is not 35%. The effect of COVID-19 on inequality is likely to be adverse and severe.

Loosely following Camus, we suggest that COVID-19 exposes the fault lines in society and amplifies inequalities. In the UK, the myth of the great equaliser has been dispelled by the publication by the Office for National Statistics (ONS) of COVID-19 mortality rates according to level of deprivation.3 It shows a clear social gradient: the more deprived the area the higher the mortality. The gradient suggests that the ‘fault line’ is not quite accurate. It is not ‘them’ at high risk and the rest of ‘us’ at acceptable risk, but a gradient of disadvantage. The argument that we are seeing COVID-19 imposed on pre-existing health inequalities is supported by the ONS figures showing that the gradient, by area deprivation, for all-cause mortality is similar to that for COVID-19.

The case that we are seeing a general phenomenon of health inequalities is shown further by a graph (figure 1) produced by the Nuffield Trust ( For shorthand, rather than the gradient, it shows mortality in the most deprived 10% and that in the least deprived 10% of areas. Remarkably, the twofold increase is consistent across a range of causes of death, including COVID-19. In the past, observing this general phenomenon, one of us (MM) speculated about general susceptibility to illness following the social gradient, perhaps linked to psychosocial processes.4 There may be elements of that. But the susceptibility may also be happening at the social level, being relatively disadvantaged puts you at higher risk of a range of specific causes of illness—the causes of the causes.

Figure 1

Mortality rate in most deprived areas.

The inequalities that the pandemic exposed had been building in the UK for at least a decade. Health Equity in England: The Marmot Review 10 Years On documented three worrying trends, since 2010: a slowdown in increase in life expectancy, a continuing increase in inequalities in life expectancy between more and less deprived areas and increased regional differences, and a decline in life expectancy in women in the most deprived areas outside London.5 The recent report examined five of the six domains that had formed the basis of the 2010 Marmot Review6: early child development, education, employment and working conditions, having at least the minimum income necessary for a healthy life, and healthy and sustainable places to live and work.

Our conclusion was that it was highly likely that policies of austerity had contributed to the grim and unequal health picture. To take just one example, highly relevant to what is happening during the COVID-19 pandemic, the crisis of adult social care. Spending on adult social care was reduced by about 7% from 2010, but in a highly regressive way. In the least deprived 20% of local authorities, the spending reduction was 3%; in the most deprived it was 16%. The UK came into the pandemic with weakened social and health services.

We drew attention to ethnic inequalities in health, but lamented that data were insufficient to give the kind of comprehensive attention we had given to socioeconomic inequalities.5 In the pandemic, the high mortality of some ethnic groups is of particular concern. There is no need, as some commentators are likely to do, to invoke genetic or cultural explanations. ONS analyses suggest that about half of the excess—in people of African, Pakistani and Bangladeshi background—can be attributed to the index of multiple deprivation.7 It may well be that this index does not capture differences in crowding that come with multigenerational households or occupational exposures.

Considering the amplification of inequalities, it is the societal response—lockdown and social distancing—that will both increase inequalities in exposure to the virus and inequalities in the social determinants of health. A most basic requirement of living in a society is that people should be able to eat. The Food Foundation’s survey reveals that 5.1 million adults in families with children have experienced food insecurity since the start of lockdown; 2 million children in those households have been food insecure (

The advice is to work from home. The lower people’s income, the less likely are they to be in jobs where working from home is possible. For example, ONS reported that before the lockdown only 10% of workers in accommodation and food could work from home; 53% of workers in communication and information could work from home. ONS showed high COVID-19 mortality in ‘front-line’ occupations such as workers in social care, drivers, chefs and sales and retail assistants.8

The paper in this issue of JECH by Fancourt and colleagues looks at experience of adversity in the UK since the start of lockdown. They show that for loss of income and employment, and for difficulties in accessing food and medicines, there is a clear social gradient—the lower the socioeconomic position the greater the adversity.

Our recent report called for a national commitment to reduce social and economic inequalities and thereby achieve greater health equity.5 As we emerge from the pandemic, such societal commitment will become ever more important.



  • Contributors The co-authors are Professor Sir MM and Dr JA from the UCL Institute of Health Equity, Department of Epidemiology & Public Health.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors. The Marmot Review 10 Years On was supported by The Health Foundation.

  • Competing interests None declared.

  • Patient consent for publication Not required.

  • Provenance and peer review Not commissioned; internally peer reviewed.

Linked Articles