18.02.2021 | Corona statistics

COVID-19 reduces life expectancy, but does not take away years of life

According to preliminary figures from Statistics Austria, the average life expectancy in Austria decreased by six months in the year 2020. But that does not mean that Austrians will live to be any less old, explains demographer Marc Luy from the OeAW.

If the pandemic is overcome soon, life expectancy in Austria should increase again.
If the pandemic is overcome soon, life expectancy in Austria should increase again. © Shutterstock

According to the available population figures, the coronavirus pandemic led to a six-month reduction in average life expectancy in 2020. Nevertheless, people will not live any shorter overall. Sound contradictory? Marc Luy from the Vienna Institute of Demography of the Austrian Academy of Sciences (OeAW) explains why it is not.

The average life expectancy is recalculated for each year. This means that a baby born today would live a total of half a year less if every year were like 2020: "For these hypothetical average babies, every birthday takes place in 2020 in the middle of the corona pandemic," explains Luy. But that is unlikely – if we manage to overcome the pandemic soon, life expectancy in Austria should increase again, the demographer explains in an interview.

A six-month decrease in life expectancy sounds dramatic. How do you rate that?

Marc Luy: The fact that life expectancy fell in 2020 is in itself no great surprise. This has been suggested by the death rates for some time. The fact that more people, especially above the age of 60, died as a result of the pandemic, has a particularly strong impact on the life expectancy indicator. A decline of six months is a noticeable kink in the trend, but also not a once-in-a-century event. Average life expectancy always shows short-term fluctuations. In 2015, we had a decrease of 0.3 years in Austria. In some neighboring countries, life expectancy even fell by half a year.

How is life expectancy even calculated?

Luy: That is a really important question. This is because the indicator is not what most people would expect based on its name. The term life expectancy sounds like a prognosis and the decline due to the corona pandemic sounds as if we would all now live half a year less than before. That is not correct.

Average life expectancy is recalculated for each year by relating all deaths in the year, broken down by age and gender, to the corresponding number of people alive. From this, one then determines for each age how high the probability was that year of experiencing the next birthday. Put simply, the current decline in life expectancy means that a newborn baby would live six months less if each of their life years looked like 2020.

So, it is a vertical section through the population structure?

Luy: Exactly. Average life expectancy connects the data from the respective calendar year and expresses how long newborn babies would live on average if they were to experience the exact survival probabilities of this year for their entire life at any age. For these hypothetical average babies, every birthday takes place in 2020 in the middle of the corona pandemic.

In fact, hopefully we will only have one or two birthdays at most during the pandemic. The purpose of the life expectancy indicator is to express the current health situation in a country as a number that can be compared with previous years. That is why the average life expectancy in countries like Austria usually increases a little every year because medical care and living conditions improve.

So, the current decline in life expectancy does not mean that people born in 2020 will, on average, live to be less old?

Luy: You can also project the actual age at death of those living today, if you want. But that is a completely different calculation, because one would estimate the survival probabilities in a longitudinal section instead of calculating them for the cross-section. That is why the term life expectancy is poorly chosen, especially in the cross-sectional view, because it arouses false expectations. However, it works well as a tool to express the mortality rates of a population in a given year.

How long will it take until the negative effect of COVID-19 on life expectancy disappears?

Luy: Events like corona leave traces in life expectancy that should disappear again as soon as the pandemic is over. But if, in addition to the current increase in mortality, there are severe long-term consequences for many of the surviving infected people, then corona effects in life expectancy could be seen in future years. For example, if it turns out that COVID-19 sufferers are more susceptible to cardiovascular diseases in 10 to 20 years. But we do not know that today. However, the short-term consequences of a pandemic on life expectancy usually subside quickly.

How strong is the influence of random fluctuations?

Luy: They can make a big difference and distort the picture of current life expectancy. One reason for the sudden drop in life expectancy in 2015 was that 2014 was an upward outlier. In 2014, the flu epidemic was very weak and there were comparatively few heat-related deaths in summer. In the spring of 2015, there was a much more severe wave of flu, which led to a greatly increased number of deaths.

The decline in life expectancy only appeared so dramatic when compared with the favorable previous year. Compared to 2013, life expectancy was higher in 2015. Life expectancy also increased significantly in 2019, although not as much as in 2014. Only a small part of the 2020 decline can therefore be attributed to outlier effects in 2019. Therefore, with a probability bordering on certainty, we can say that the coronavirus is responsible for most of the additional deaths and the decline in life expectancy.

Does that mean that life expectancy in Austria will increase sharply in 2021?

Luy: Life expectancy will probably rise again in 2021 if we quickly win the fight against the pandemic and it does not continue to cause additional deaths. It is difficult to predict how strong the increase will be. That depends on how quickly we get the pandemic under control and what else happens during the year. However, if we do not manage to contain the virus and mortality remains high, life expectancy could also remain at the current level or maybe even decrease a little.

Has the coronavirus also had positive effects on life expectancy, for example through a lower number of traffic fatalities?

Luy: It is entirely possible that there will be a lower number of traffic fatalities because of the lockdowns. For the value of life expectancy, however, this decline will be less significant than the increase in deaths from the virus. In addition, corona probably also causes indirect deaths, for example if treatment for other diseases such as cancer is started later due to medical bottlenecks.

Are the numbers for 2020 already fixed or are they still estimates?

Luy: Complete data are not yet available for 2020. However, the estimates of life expectancy published by Statistics Austria are unlikely to change much. In the coming months, more detailed studies on the influence of corona will certainly appear – these will also examine the effects on individual groups, for example according to age, gender or education. We expect that decreasing life expectancy will affect men, the elderly and people from socially weaker groups more. At least some international studies suggest this. For Austria, however, this is still speculation. When we have the detailed data for 2020, especially on the causes of death, we can take a close look.



Marc Luy heads the research group “Health and Longevity” at the Vienna Institute of Demography of the Austrian Academy of Sciences (OeAW) and is a corresponding member of the OeAW. Before that he was Junior Professor for Demography at the University of Rostock. He has conducted research at, among others, the Max Planck Institute for Demographic Research in Rostock and the Federal Institute for Population Research in Wiesbaden, Germany.