When will the COVID-19 pandemic end?

July 28, 2022

The past few months have seen many parts of the world, including Europe and North America, continue their journey toward endemic COVID-19. While the BA.5 subvariant has produced a rise in the number of cases in many places, the burden of severe disease remains low in Europe and is only moderately higher in the United States, thanks to the lower average severity of Omicron subvariants and high levels of partial immunity.1 Countries such as Australia and New Zealand have decreased their previously strong focus on controlling viral spread and are now transitioning to strategies for managing endemic disease. They have seen continued success, recording higher case counts but few deaths. The experience of these countries, with total excess mortality from the pandemic far lower than it is in other parts of the world and only a limited need for ongoing restrictions, can make a case that their response to the COVID-19 pandemic has been among the most effective in the world.2

As of mid-July 2022, mainland China is the only large country pursuing a zero-COVID-19 strategy.3 That approach has kept the number of deaths caused by COVID-19 low. Nonetheless, the relatively low levels of immunity among the population, including the elderly, make the endgame for this approach less clear than it is for countries that have shifted away from a zero-COVID-19 strategy. In the meantime, the current approach appears to have had substantial and mostly negative effects on both the Chinese and the global economies.4

Looking ahead, the outlook for most regions, including Europe and North America, remains relatively favorable for the coming months, with levels of severe disease at or below recent levels. The winter of 2022–23 may see a more substantial uptick in the Northern Hemisphere, but this is unlikely to be as severe as the December 2021–February 2022 wave. Two caveats are important. First, the term “relatively favorable” needs qualification. While many people in the United States are growing comfortable living alongside COVID-19, the average number of daily deaths still runs at two to four times the long-term average for influenza, and it’s higher on a seasonally adjusted basis.5 Second, a new, immunity-evading variant could upend this relatively optimistic story at any time. Delta-cron (Omicron’s twin) and Milder-cron scenarios we described in our March 2022 update remain a reasonable summary of the range of possibilities for the impact of future variants.

Subvariants of Omicron, especially BA.5, have challenged the world with ever more transmissible versions of the virus. But so far, they haven’t fundamentally changed the dynamics of the pandemic, because there hasn’t been a step change decline in immunity, as seen during the winter, when Omicron first emerged. To be sure, some tentative evidence that BA.4 and BA.5 can evade immunity is emerging.6 For the moment, however, scientists view these variants as simply more transmissible than Omicron. The vaccines now in wide distribution still limit the incidence of severe disease.

Vaccines, therapeutics, and tests

Although advances in detecting and treating COVID-19 lead the news less frequently than they did a year ago, important developments continue to emerge. Public-health authorities around the world are considering short- and medium-term strategies for the timing and rollout of booster shots. Many are already recommending fourth doses for those at highest risk of severe disease, given recent studies highlighting the additional benefit provided.7 Some are debating a strategy that would combine annual flu and COVID-19 boosters in the autumn.8

However, the future uptake of boosters is a significant unknown. In the United States, only about half of those who received the initial course of COVID-19 vaccines have received the first booster. It isn’t yet clear whether public interest in boosters will continue to decline or demand will in time match the historical uptake of flu vaccines (around 50 percent of adults). Without continued immunization, population immunity to current variants, including Omicron, will lessen over time. But Omicron-specific boosters (especially to address BA.4 and BA.5) could potentially emerge this autumn and significantly enhance the efficacy of vaccines against the predominant strain.9

The recent authorization of COVID-19 vaccines for children as young as six months is also an important step.10 Of course, the low rate of severe COVID-19 in this age group means that the impact on pediatric mortality will be small. But there is hope that vaccinating young children will increase overall population immunity (a critical metric, as we discuss later in the article).

Paxlovid and other COVID-19 therapeutics are now widely available in high-income countries. In the United States, supply is well ahead of demand.11 Increasing the uptake of effective therapeutics is an important step for governments as they continue the transition toward managing endemic COVID-19. These therapeutics, including both small-molecule and antibody treatments, have helped blunt the incidence of severe disease in places where they are widely available. But the same shortfalls in access that bedevil the distribution of vaccines in low-income countries are striking again with therapeutics: doses sit unused in high-income countries while other parts of the world lack access.12

Testing for COVID-19 has increasingly shifted to a home-based model. Since so few results from home tests are reported, it has been difficult to estimate the current wave’s true magnitude in the United States. The official statistics, which already tend to underestimate the number of cases, may become even less sensitive to the true dimensions of viral transmission.

Public-health authorities may therefore need to look at a wider range of indicators to support their planning and disease management efforts. One such indicator may be consumer behavior. Since home tests became more widely available, in late 2021, consumer purchases of them have risen in line with waves of COVID-19 cases, prefiguring rises in deaths attributable to the disease (Exhibit 1). It’s too soon to tell if rising numbers of test purchases reflect consumers’ awareness of symptomatic disease (when family members fall ill, for example), their knowledge of the spread of cases in the community, or other factors. But it seems likely that purchases of tests (like online searches of “flu near me”) are an indicator and could provide some advance warning of future waves of the disease. Over the past couple of months, the number of tests bought from one online retailer has been rising, which may reinforce the impression that the United States is entering a period of sustained case growth.

Purchases of tests seem closely connected to case counts.

Estimating community risk

The risk that COVID-19 poses to a community at any point in time is a function of the immunity of its members. Immunity protects individuals and reduces risk for people in the community around them by reducing rates of onward transmission. It is gained through vaccination (primary series and boosters), infection with SARS-CoV-2, or both (hybrid immunity). It is lost both gradually, with the passage of time, and suddenly, when a new variant that evades the immunity afforded by vaccination or previous infection becomes dominant (as happened with the emergence of Omicron in late 2021). A community’s level of immunity is a precarious balance, constantly shifting as individuals gain and lose immunity.

Although each individual’s level of immunity cannot be measured in real time, we can base estimates of the level of community risk on what is known about vaccination rates and previous infections. Exhibit 2 summarizes that knowledge, based on the published literature and values derived from it. Note that the values shown define ranges of immune protection against symptomatic infection, since that is the metric most of the published literature uses. Protection against any infection (including asymptomatic disease) is likely to be lower—and protection against severe disease is likely to be higher. Ranges reflect the uncertainty around immunity levels and describe population averages. They are not intended to be predictive for any individual.

The McKinsey COVID-19 Immunity Index uses values drawn from the medical literature.

This information, combined with estimates of when infections and vaccinations took place, enables us to create the McKinsey COVID-19 Immunity Index, showing fluctuations in degrees of protection over time (see sidebar, “Description of methods for the McKinsey COVID-19 Immunity Index”). Exhibit 3 includes three types of immunity for six countries: immunity derived from previous infection only, from vaccination only, and from both (now the largest group in many countries). Higher scores mean that more people have immunity and that the community has greater protection against symptomatic disease.

National immunity levels have varying components and change constantly.

Two communities might have the same score but different combinations of vaccination and prior infection. They might also have different outlooks for the next few months if their collective immunities are waning quickly or slowly. Note that immunity against symptomatic disease (not any disease) is what we try to assess; we don’t estimate a community’s threshold for herd immunity, as that would be driven by both symptomatic and asymptomatic infections.

This initial version of the McKinsey COVID-19 Immunity Index helps us make a few observations:

  • The emergence of Omicron during the winter of 2021–22 is visible as a sharp drop in immunity in multiple countries (since existing immunity was suddenly less effective against the new variant).
  • The proportion of unvaccinated people with past infections in a country is roughly correlated with its overall COVID-19 mortality, since unvaccinated but infected people have been at greatest risk of bad outcomes.13
  • Even countries, such as Italy and the United Kingdom, with strong vaccination track records may be starting to lose immunity as the pace of booster uptake drops over the summer (although most countries are prioritizing boosters for the highest risk populations).
  • As Australia transitions its COVID-19 strategy, it has rapidly gained immunity over the past year through a combination of vaccination and infections.
  • Vaccination remains an essential element in building immunity. Some new studies suggest that infection confers more immunity and protection against severe disease than does vaccination alone.14 As a result, some countries with strong vaccination records may now have lower immunity than others, such as the United States, that have experienced a large number of cases during recent Omicron waves and for which much of the population was last immunized about six months ago.
  • Even countries, such as the United States, with high McKinsey COVID-19 Immunity Index scores continue to experience many new COVID-19 cases as a result of the extremely high transmissibility of current SARS-CoV-2 variants. US mortality for COVID-19 is still higher than the historical average for flu but has fallen by 85 to 90 percent since its early 2021 peak.15
  • A successful autumn booster campaign in Northern Hemisphere countries will be important to increasing immunity levels before winter comes.

One significant limitation of the McKinsey COVID-19 Immunity Index is that it doesn’t indicate which people within a community are protected. A country in which the elderly are overwhelmingly immune will have much better outcomes than will a similar country where the same level of overall immunity is concentrated in the young. Immunity can only partly predict the burden of disease at any point in time. Other important drivers of disease trends include the variant mix, the season, and behavior (such as masking, compliance with isolation and quarantines, and working from home). However, the McKinsey COVID-19 Immunity Index can be helpful in understanding a society’s level of protection. Measurements like it can help inform both individual behavior and public policy during the next chapter of the COVID-19 pandemic.

About the authors

Sarun Charumilind is a partner in McKinsey’s Philadelphia office, Matt Craven is a partner in the Silicon Valley office, Jessica Lamb is a partner in the New Jersey office, Adam Sabow is a senior partner in the Chicago office, Shubham Singhal is a senior partner in the Detroit office, and Matt Wilson is a senior partner in the New York office.

The authors wish to thank Giles Colclough, Alina Glukhonemykh, Abhishek Sharma, and Zihao Xu for their contributions to the article.

This article was edited by Mark Staples, an editorial director in the New York office.

March 1, 2022

After the short, sharp shock of Omicron, the pandemic phase of COVID-19 looks to be ending for most locations, unless a significant and severe new variant emerges. This update discusses what we’ve learned from Omicron, the prospects for the rest of 2022, and presents three potential criteria for defining COVID-19 as endemic.

Since the Omicron variant of COVID-19 was named by WHO on November 26, 2021, it has moved at lightning speed. In less than three months, Omicron has spread around the world, caused record peaks in cases in many places,16 and is now declining just as quickly. In most places, the worst of the Omicron wave has passed leading some locations to loosen public-health measures to a degree not seen in almost two years. In contrast, some locations, such as Hong Kong,17 are seeing their worst peak yet, and continuing to tighten restrictions.

The most likely scenario we discussed in the December 2021 edition of this article has proved to be largely accurate—Omicron is more infectious than any previous variant and evades the immunity provided by both prior infection and incomplete vaccination18 (Exhibit 1). These factors, combined with limited behavior change from pandemic-weary populations—and the twin accelerants of transmission, holiday travel and gatherings—meant that Omicron moved through the population with remarkable speed. Fortunately, the early evidence that Omicron is, on average, less severe than Delta also proved right.19 The worst-case scenarios were avoided. The even more infectious BA.2 sub-variant of Omicron may have worsened the wave but has not substantially changed this narrative to date.20

Omicron is more infectious than other common viruses, and less fatal than Delta.

What we’ve learned

The Omicron wave has taught us several lessons about the effectiveness of various societal responses. First, up-to-date vaccination status, including a recent booster, proved to be especially important in protecting against Omicron212223. Countries where a significant portion of those at risk had received three doses of vaccine, including at least one dose of mRNA vaccine, saw hospitalizations substantially decouple from cases.24 This meant that many European countries had more cases but fewer hospitalizations during this wave than prior ones. On the other hand, locations with lower up-to-date vaccine coverage, including parts of the United States, set all-time records for hospitalization and deaths. As in previous waves, lower-income countries and those with younger populations were somewhat protected,25 even though inequalities in global vaccine access have meant that few there have received three doses, and most have not yet received a single dose.

Second, the link between cases and behavioral adjustments is largely broken. Data shows that more and more people have concluded that the health risks of COVID-19 are not significant enough for them to change their behavior, either because of their vaccination status, their youth, or a desire to move on from the pandemic.26 Third, and consistent with this trend, some governments have concluded that the total societal costs of lockdowns, restrictions on business, or masking outweigh the benefits at this phase of the pandemic.2728 Other governments, however, are maintaining or strengthening public-health policies, including vaccine mandates.29 Many workplaces remain relatively cautious in their policies,30 but public-health responses to Omicron have typically been less forceful than those of prior waves with similar disease burdens.

The next ten months

Prospects for the rest of the year and beyond hinge on the questions of whether and when future variants will emerge. As long as Omicron remains the dominant variant, there is reason for relative optimism. Our scenario analysis suggests that Omicron-related hospitalizations are likely to continue to decline in the United States and remain at relatively low levels through the spring and summer (Exhibit 2). We might then expect to see a seasonality-driven wave of disease next fall and winter, but hospitalizations would likely peak well below the level of the wave we just experienced.

If Omicron remains the dominant variant, US hospitalizations will likely stay low throughout 2022.

The default scenario, in which Omicron remains the dominant variant, represents a continuation of the transition toward managing COVID-19 as an endemic disease that is already underway in many locations. With Omicron as the dominant variant, the pandemic phase will feel like it is over for more and more people, though certainly not all.

As ever, different parts of the world will experience the coming phase differently. Countries with high rates of current immunity and widespread booster uptake will be better protected. Age demographics will continue to be an important risk driver. The dynamics of seasonality may cause differences between the northern and southern hemispheres. And government policy still matters—in particular, the few remaining countries with zero-COVID-19 strategies may also experience the coming months differently as they choose whether to continue or relax their border policies.

New variants: the big unknown

By and large, the six-month outlook in many countries is brighter than at any time in the past two years. But several uncertainties could temper the optimism, starting with the duration of immunity. Evidence suggests that both natural and vaccine-induced immunity wane over time, particularly against infection.31 While we don’t yet know the full extent of waning immunity for Omicron, new evidence indicates that those who have received three doses of vaccine may benefit from medium-term protection.32 At the same time, booster uptake has been significantly lower than first- and second-dose coverage in many countries. For example, while 215 million Americans are fully vaccinated, only 93 million have also received a booster dose.33 So, as we consider future waves, two critical questions remain about the duration of protection: how significantly will immunity wane? And will booster uptake continue to slow in each subsequent round of boosting?

The next wave of medical advances will also prompt questions. Pfizer and Moderna have indicated that modified vaccines targeted against Omicron could be available in the coming months,3435 but we don’t yet know their efficacy, duration of protection, or the policies that will be set around fourth doses. Nor is it clear yet what the approval standards might be for multi-valent vaccines. On another front, there is hope that wider use of the oral therapeutics paxlovid and molnupiravir will further decrease the number of severe cases,3637 but the real-world impact of their use at scale is not yet known, and supplies of paxlovid are still scaling.38

While these uncertainties are important, they do not necessarily change the story of a transition toward endemicity under Omicron. The main risk to that transition is a significantly different new variant that replaces Omicron as the dominant strain. We made this point in the last four editions of this article, and unfortunately it remains as true as ever. SARS-CoV-2 will continue to mutate under all scenarios, but most mutations do not lead to stable new forms of the virus with an evolutionary advantage. Alpha, Delta, and Omicron have met this standard, and have changed the trajectory of the pandemic. Beta and Gamma have also affected the trajectory, but to a lesser degree; their evolutionary advantage was not great enough to become globally dominant.39

Omicron is already among the most infectious human viruses known to science.40 While even greater infectiousness (such as the sub-variant BA.2 has exhibited) is possible, to become dominant a new variant would likely need to also partially or fully evade prior immunity, including that provided by Omicron infection. If such a variant emerged, its average clinical severity would then be critical. Exhibit 3 lays out three example scenarios for the potential characteristics and trajectory of the pandemic under a new dominant variant. This is not a complete list of possible future variants but some potential options. (Note: these scenarios are not related to the Omicron-hospitalization scenario shown in Exhibit 2.)

Here are three example scenarios for a new dominant variant of COVID-19.

A new dominant variant will receive a Greek-letter name, but until then we have created more descriptive names for our scenarios. Under the “Omicron’s twin” scenario, a variant that evaded prior immunity (including from Omicron) but was otherwise similar to Omicron in transmissibility and severity of disease might cause a wave of disease broadly similar to the one we have recently experienced, though perhaps slightly worse if the public response to it is even more muted and if vaccine-conferred immunity has waned. A worse case might be “Delta-cron”, a variant that evades prior immunity and combines the infectiousness of Omicron with the average severity of Delta. This might occur if vaccines proved less effective in preventing severe disease, and could lead to the worst wave yet for many locations. The “Milder-cron” scenario would continue the trend toward less severe disease. Countries might then experience a smaller version of the recent Omicron wave, which might be managed similar to the way societies manage flu on an ongoing basis.

What’s more difficult to estimate is when a new variant of SARS-CoV-2 will emerge. It could be a day after we publish this update, or six months , or years from now. The extraordinary progression we’ve already seen—in just over two years four strains in succession have become globally dominant—makes it dangerous to plan on a “no new variant” scenario. But it is possible that evolution will not produce epidemiologically significant new variants. The risk of new variants emerging is related to the number of cases in the world, since each infected individual represents a new opportunity for viral evolution. For this reason, the continued global rollout of COVID-19 vaccines remains an investment in our collective safety as well as an imperative to protect individuals.

Some have suggested that particular populations, such as those who are immunocompromised due to HIV or other causes, are disproportionately at risk of incubating new variants.41 Others have posited a possible zoonotic origin of Omicron.42 As scientific understanding of these potential pathways develops and genomic surveillance networks continue to expand, societies may get better at reducing the risk of variant emergence. For now, we remain bystanders as the virus evolves.

Pandemic to endemic: Where does one end and the other begin?

Several potential definitions of the transition from pandemic to endemic phase are possible (Exhibit 4). Epidemiologically, COVID-19 can be defined as endemic when it exists at a predictable level that does not require society-defining interventions.43 While we all wish that level could be zero, eliminating the disease is not feasible for any country with open borders. Previous editions of this article invoked a comparison of the COVID-19 burden to that from other diseases such as flu. Just as the risk of flu is considered normal, so too might the risk of COVID-19. In addition, what is considered acceptable by society will differ across countries. Countries currently reopening during Omicron wave downswings are doing so amid very different experiences of COVID-19 burden. For example, the COVID-19 death rate per capita in the past month for the United States is 50 percent higher than Argentina’s and ten times greater than the Philippines’.44

Three distinct definitions for COVID-19 endemicity are emerging.

A behavioral threshold for endemicity would come when fluctuations in disease burden cause only minimal change in individuals’ economic and social behavior. This is mediated by individual risk factors (age, underlying conditions, and so on), and their risk appetite.

Finally, an economic threshold for endemic COVID-19 will come when epidemiology substantially decouples from economic activity and secondary economic effects largely resolve. This economic definition is related to the individual behavior definition, but may take longer to reach because those secondary effects, including supply chain imbalances, labor market disruptions, and global asymmetries affecting travel and trade, may linger.

Until a new variant emerges, and under some scenarios even once it does, the United States and Europe will likely continue to move toward these definitions of endemicity. As that happens, countries across Europe are rolling back the last public-health restrictions.45 England plans to end the isolation requirement for those testing positive (and is ending free asymptomatic testing).46 The US CDC recently announced a change in its masking guidelines that serves to significantly reduce the number of areas where masking is recommended.47 Airline passenger volumes in the United States are much closer to prepandemic levels than they were a year ago48 and schools navigated the recent wave with less disruption than was caused by previous waves of disease.49

Endemic COVID-19 does not mean that the disease poses no risk. Globally, we should aim for an “always on” response system that can scale quickly. And as we have written previously, every society must do four things to manage COVID-19 effectively during the endemic phase:

  • Choose a holistic set of health, economic, and social markets that they are managing for
  • Monitor and track progress against them in ways that allow for targeted response escalation when needed
  • Limit disease through effective use of vaccines, therapeutics, and other countermeasures
  • Slow transmission through testing and environmental/workplace modifications

A new variant may yet trigger another chapter in the COVID-19 pandemic and societies must be prepared to respond if and when that happens. But for now, the pandemic phase looks to be ending.

About the authors

Sarun Charumilind is a partner in McKinsey’s Philadelphia office, Matt Craven is a partner in the Silicon Valley office, Jessica Lamb is a partner in the New Jersey office, Adam Sabow is a senior partner in the Chicago office, Shubham Singhal is a senior partner in the Detroit office, and Matt Wilson is a senior partner in the New York office.

The authors would like to thank Richard Lu and Aurora Xu for their contributions to this article.

This article was edited by Mark Staples, an executive editor in the New York office.

December 15, 2021

The Omicron variant is spreading rapidly. What does it hold in store? Our analysis offers some scenarios to understand potential outcomes. We also look at the effects of boosters, the potential waning of vaccine efficacy, and new oral therapeutics.

On November 26, 2021, WHO reached deeper into the Greek alphabet to declare Omicron a new SARS-CoV-2 variant of concern.50 The world’s reaction has been an unpleasant mixture of dread, fatigue, and déjà vu. Almost two years into a pandemic that has claimed more than five million lives and affected billions more, people everywhere are finding it hard to summon the energy for another chapter in the story.51

Endemicity remains the endpoint. But at the time of writing, the Omicron variant is rewriting the timetable. Whether because Omicron is more infectious or has greater ability to evade the immune system, or both, it quickly became the dominant variant in South Africa.52 Data so far are mixed on the severity of the disease it causes: some early findings have pointed toward a mild clinical course, while other evidence has suggested that Omicron may lead to more frequent hospitalization in children than other variants do.53 We have written previously about the transition to managing COVID-19 as an endemic disease and noted that a new variant was one of the greatest risks to timelines.

This article presents a new analysis of a range of scenarios based on the infectiousness, immune evasion, and severity of disease caused by the Omicron variant. Based on the evidence to date, we have posited a base-case scenario in which Omicron is about 25 percent more infectious, evades prior immunity to a greater degree (25 percent), and causes less severe disease, again by about 25 percent, all relative to Delta.54 Our analysis suggests that in the United States, this combination of characteristics would lead to Omicron replacing Delta as the dominant variant in the next few months and to a higher peak burden of disease than the country saw in the second half of 2021 (but likely below the peak reached in the winter of 2020–21).

This base-case scenario has the potential to place a severe strain on healthcare systems. The optimistic scenario would see a peak of disease burden close to that seen over the past six months, while the pessimistic would see a very significantly higher burden of disease than in the past six months. Note that in every scenario, our analysis indicates that hospitalizations will likely be higher in the next six months than they were in the past six months.

In any scenario for the future of the COVID-19 pandemic, much depends on the ways in which societies respond. Three levers are likely to be especially important, starting with the extent to which countries can effectively scale and make available new oral therapeutics with the potential to reduce the chance of progression to severe disease, and which are unlikely to be blunted by Omicron. Second, evidence is accumulating that booster doses are especially important for protecting against the Omicron variant; accelerating their rollout will help protect populations. And third, given public fatigue and the lessons of the past two years, finding the right combination of public-health measures will be critical.

The Omicron variant

Three main factors determine the real-world impact of any new SARS-CoV-2 variant: the extent to which it can evade the immunity developed by those who have been vaccinated or previously infected by other variants, its inherent infectiousness (often expressed as a higher basic reproduction number, or R0), and the severity of disease caused.55 The first two factors combine to drive the number of cases, while the third determines the number of severe cases and deaths. For example, the Delta variant, which remains dominant in most of the world, was significantly more transmissible than previously circulating variants were, showed limited incremental evasion of immunity, and caused moderately more severe disease relative to other variants.56

Early data paint a mixed picture of Omicron’s evasion of vaccine-induced immunity. The UK Health Security Agency recently summed up its view: “Early estimates of vaccine effectiveness (VE) against symptomatic infection find a significantly lower VE [against] Omicron infection compared to Delta infection. Nevertheless, a moderate to high [VE] of 70 to 75% is seen in the early period after a booster dose.”57

There is still much more to learn—sample sizes in the new studies were small, antibody titers are an imperfect metric of immune protection, and major manufacturers are yet to release similar information. The response to Omicron may include both accelerating the rollout of booster doses of existing vaccines and developing new formulations better targeted to this variant. Companies have indicated that modified or new vaccines could be available in a few months, though the scale and global availability are unclear.58

Regarding evasion of natural immunity, a preprint article from South Africa suggests a significantly higher chance of reinfection by Omicron relative to Delta or Beta.59 Both the pace of case growth and the rapidly increasing share of Omicron among samples sequenced suggest that, through a combination of greater infectiousness and immune evasion, Omicron is spreading very quickly.60 If the experience of South Africa were to be repeated elsewhere, we could see a continued rapid increase in the number of COVID-19 cases as Omicron is established.

The question of disease severity is more complicated. Several clinicians in South Africa have noted the apparently mild presentation of Omicron cases.61 Further, the European Centre for Disease Prevention and Control (ECDC) noted on December 12, 2021, that 776 cases were within its remit and “all cases for which there is available information on severity were either asymptomatic or mild. There have been no Omicron-related deaths reported thus far.”62 On the other hand, ECDC also notes that it is too early to draw definitive conclusions on disease severity. The United Kingdom reported its first Omicron-related death on December 13, 2021, and some reports from South Africa suggest a potentially higher rate of hospitalization among young children than seen in previous waves of COVID-19.63

In the base-case scenario, US COVID-19-related hospitalizations could peak significantly higher in the next six months than in the past six months.

Each of these observed trends may change as sample sizes increase, confounding factors are considered, and the clinical course of disease plays out over time. The answers, when they arrive, will have important consequences for the months ahead. Given the uncertainty, we have built a set of scenarios describing potential outcomes measured by hospitalization rate. They are indexed on the recent Delta wave and show whether various potential combinations of infectiousness, immune evasion, and clinical severity are likely to lead to a higher or lower rate of COVID-19-related hospitalization.

The results of these scenarios for the United States are shown in Exhibit 1. Each of the three variables is an important driver of the outcomes. Evidence so far suggests that the Omicron variant, relative to Delta, is likely to be more infectious, show more immune evasion, and be less severe, on average. In the base case (25 percent more infectious; 25 percent greater immune evasion; 25 percent less severe disease), the COVID-19-related hospitalization rate in the United States could peak significantly higher in the next six months than in the past six. In the pessimistic scenarios, the peak number of hospitalizations for COVID-19 could be much higher in the next six months than in the past six months, whereas in the optimistic scenario, the number would be higher but similar to that seen in the second half of 2021, as waning immunity causes ongoing disease from a combination of the Delta and Omicron variants.

The analysis is quite sensitive to public-health interventions and behaviors. Exhibit 1 assumes a US public-health response similar to that seen during the Delta wave. Exhibit 2 shows potential outcomes if more stringent public-health measures were to be employed in the United States; this could lead to a disease burden that is similar to or only moderately worse than seen in the past six months, depending on the characteristics of the virus. The exhibit also shows how much more stringent those measures would need to be to potentially prevent the disease burden from exceeding the burden of Delta. Our analysis accounts for waning immunity and suggests that even if Omicron were to have no impact, the next six months of Delta-driven disease in the United States could be about as severe as the past six months were.

If Omicron is more infectious than Delta, and society adopts significant new interventions, the increase in COVID-19 hospitalizations may be moderate.

In the context of Omicron’s arrival and impending spread, three factors come to the forefront: the potential impact of new therapeutics in reducing hospitalizations and death, the criticality of boosters in the context of waning immunity, and clarity and consensus in public-health measures.

Oral therapeutics

The clinical management of COVID-19 has come a long way since the early days of the pandemic. The availability of effective monoclonal antibodies, dexamethasone, and other treatments and the use of nonpharmacological interventions, such as “proning,” have meaningfully increased the chances of survival for those with access to high-quality healthcare.64 Nevertheless, recent results from Merck–Ridgeback Biotherapeutics and Pfizer on their oral drugs molnupiravir and PAXLOVID, respectively (two antivirals, with different mechanisms of action65) represent a material advancement and increase the chance that the impact of the Omicron variant can be controlled. In its final study, Pfizer reported that PAXLOVID reduced risk of hospitalization or death by about 89 percent for high-risk patients and about 70 percent for standard-risk patients.66

Oral therapeutics that significantly reduce the chance of progression to severe disease after symptom onset may enable a higher fraction of cases to be managed as outpatients. Such therapies are also easier to administer in lower-resourced regions than injected or infused treatments are. Further, manufacturing small molecules rapidly is faster than the process for monoclonal antibodies. The initial evidence indicates that the efficacy of these therapies is unlikely to be reduced by the mutations present in the Omicron variant.67

A number of questions and caveats remain. Data are not yet available on the drugs’ efficacy in vaccinated individuals. The drugs are more likely to be effective if taken within five days of symptom onset,68 requiring an efficient pathway from diagnosis to prescription and distribution. And some data inconsistencies have emerged—for example, it is unclear why molnupiravir’s efficacy in an interim analysis dropped in the final readout.69

Other questions relate to the impact of the new therapeutics in blunting an Omicron-driven wave of disease. Can healthcare systems diagnose COVID-19 and distribute therapeutics fast enough for them to be effective? Will oral therapeutics be available quickly enough to blunt a potential Omicron surge in December 2021 and January 2022? How will drug–drug interactions with ritonavir be managed for PAXLOVID use? Can mutagenicity concerns in pregnant women for molnupiravir be managed to ensure patient safety while maximizing effective use of the drug? What role will antibody treatments play?

Waning immunity and boosters

The decline in the efficacy of COVID-19 vaccines over time and the benefits of booster doses have become much better understood over the past three months. While an initial course of all WHO-approved vaccines continues to provide strong protection against severe illness and death, the rate of breakthrough cases increases meaningfully as time passes, indicating that protection declines with time. For example, a July 2021 study of the Pfizer–BioNTech vaccine in Israel showed that in every age group studied, those who had been vaccinated by January 2021 were more likely to experience breakthrough infection than those who completed their initial course of vaccination two months later were.70 This general point appears to be especially true for the Omicron variant.71

Evidence has also accumulated steadily about the benefit of booster doses, leading more countries to expand and accelerate their rollout.72 On October 21, 2021, Pfizer–BioNTech announced results from a randomized controlled trial of third doses of its COVID-19 vaccine; protection was restored to the levels seen in earlier trials after the second dose.73 More recent data, as previously described, highlight the benefits of booster doses in protecting against the Omicron variant.

WHO and others have raised important concerns about the appropriateness of high-income countries offering booster doses of COVID-19 vaccines while so many in the world have not received initial vaccination, but the benefits of a booster dose to an individual patient are increasingly clear.74 The ongoing Delta-driven wave of cases in Europe has led a number of countries to accelerate their booster-dose rollout, with some discussing the timing of potential additional doses.75

As countries transition over time to managing COVID-19 as an endemic disease, the world may reach a long-term state of disease prevention similar to that seen with the flu, with annual or twice yearly booster doses. In the short term, an accelerated rollout of booster doses of COVID-19 vaccines is likely to be one of the best protections against an Omicron-fueled wave of the disease.

Public-health measures

Even before the emergence of Omicron, the past four months have seen the continued evolution of the public response to COVID-19. Debates have continued about the role of vaccine mandates, the use of vaccine passports, testing requirements, masks and mask mandates, and restrictions on gatherings. Societies are trying to find a new consensus through this transition, with some maintaining minimal public-health restrictions in the face of rising case counts and others reinstating more stringent measures. The emergence of Omicron led to tighter rules around travel in many countries, with some, such as England, also restricting domestic travel.76 Achieving some degree of consensus on public-health measures will likely be an important step toward controlling an Omicron-driven wave of disease.

Omicron is a sobering reminder that SARS-CoV-2 has the advantage of rapid mutation and can produce new variants faster than anyone would like. We hope that this article offers a starting point to interpret the potential spread and severity of the disease it produces and the ways in which new therapeutics, booster doses of vaccines, and public-health measures will be important in limiting its impact.

About the authors

Sarun Charumilind is a partner in McKinsey’s Philadelphia office, Matt Craven is a partner in the Bay Area office, Jessica Lamb is a partner in the New Jersey office, Adam Sabow is a senior partner in the Chicago office, Shubham Singhal is a senior partner in the Detroit office, and Matt Wilson is a senior partner in the New York office.

The authors wish to thank Alizeh Hasham Gangji, Giulio Morina, Konstantinos Tsakalis, and Aurora Xu for their contributions to this article.

This article was edited by Mark Staples, an executive editor in the New York office.

August 23, 2021

This article updates our perspectives on when the coronavirus pandemic will end to reflect the latest information on vaccine rollout, variants of concern, and disease progression. Among high-income countries, cases caused by the Delta variant reversed the transition toward normalcy first in the United Kingdom, during June and July of 2021, and subsequently in the United States and elsewhere. Our own analysis supports the view of others that the Delta variant has effectively moved overall herd immunity out of reach in most countries for the time being. The United Kingdom’s experience nevertheless suggests that once a country has weathered a wave of Delta-driven cases, it may be able to resume the transition toward normalcy. Beyond that, a more realistic epidemiological endpoint might arrive not when herd immunity is achieved but when COVID-19 can be managed as an endemic disease. The biggest overall risk would likely then be the emergence of a significant new variant.

Since the March installment in this series, many countries, including the United States, Canada, and those in Western Europe, experienced a measure of relief from the COVID-19 pandemic77 when some locales embarked on the second-quarter transition toward normalcy that we previously discussed.78 This progress was enabled by rapid vaccine rollout, with most Western European countries and Canada overcoming their slower starts during the first quarter of 2021 and passing the United States in the share of the population that is fully immunized.79 However, even that share has been too small for them to achieve herd immunity, because of the emergence of the more transmissible and more lethal Delta variant80 and the persistence of vaccine hesitancy.81

Among high-income countries, cases caused by the Delta variant reversed the transition toward normalcy first in the United Kingdom, where a summertime surge of cases led authorities to delay lifting public-health restrictions, and more recently in the United States and elsewhere. The Delta variant increases the short-term burden of disease, causing more cases, hospitalizations, and deaths.82 Delta’s high transmissibility also makes herd immunity harder to achieve: a larger fraction of a given population must be immune to keep Delta from spreading within that population (see sidebar, “Understanding the Delta variant”). Our own analysis supports the view of others that the Delta variant has effectively moved herd immunity out of reach in most countries for now,83 although some regions may come close to it.

While the vaccines used in Western countries remain highly effective at preventing severe disease due to COVID-19, recent data from Israel, the United Kingdom, and the United States have raised new questions about the ability of these vaccines to prevent infection from the Delta variant.84 Serial blood tests suggest that immunity may wane relatively quickly. This has prompted some high-income countries to start offering booster doses to high-risk populations or planning for their rollout.85 Data from the US Centers for Disease Control and Prevention also suggest that vaccinated people who become infected with the Delta variant may transmit it efficiently.86

These events and findings have raised new questions about when the pandemic will end. The United Kingdom’s experience nevertheless suggests that once a country has weathered a Delta-driven wave of cases, it may be able to relax public-health measures and resume the transition toward normalcy.87 Beyond that, a more realistic epidemiological endpoint might arrive not when herd immunity is achieved but when countries are able to control the burden of COVID-19 enough that it can be managed as an endemic disease. The biggest risk to a country’s ability to do this would likely then be the emergence of a new variant that is more transmissible, more liable to cause hospitalizations and deaths, or more capable of infecting people who have been vaccinated.

Raising vaccination rates will be essential to achieving a transition toward normalcy. Vaccine hesitancy, however, has proven to be a persistent challenge, both to preventing the spread of the Delta variant and to reaching herd immunity.88 The US Food and Drug Administration has now fully approved Pfizer’s COVID-19 vaccine, and other full approvals may follow soon, which could help increase vaccination rates.89 Vaccines are also likely to be made available to children in the coming months,90 making it possible to protect a group that comprises a significant share of the population in some countries.

In this article, we review developments since our March update, offer a perspective on the situation and evidence as of this writing, and present our scenario-based analysis of when a transition toward normalcy could occur.

Even without herd immunity, a transition toward normalcy is possible

We have written previously about two endpoints for the COVID-19 pandemic: a transition toward normalcy, and herd immunity. The transition would gradually normalize aspects of social and economic life, with some public-health measures remaining in effect as people gradually resume prepandemic activities. Many high-income countries did begin such a transition toward normalcy during the second quarter of this year, only to be hit with a new wave of cases caused by the Delta variant and exacerbated by vaccine hesitancy.

Indeed, our scenario analysis suggests that the United States, Canada, and many European countries would likely have reached herd immunity by now if they had faced only the ancestral SARS-CoV-2 virus and if a high percentage of those eligible to receive the vaccine had chosen to take it. But as the more infectious Delta variant becomes more prevalent within a population, more people within that population must be vaccinated before herd immunity can be achieved (Exhibit 1).

Vaccine hesitancy makes it all the more difficult to reach the population-wide vaccination level rates that confer herd immunity. Researchers are learning more about differences among individuals’ attitudes, which include both “cautious” and “unlikely to be vaccinated.” 91 Meanwhile, social tolerance for vaccination incentives and mandates appears to be growing, with more European locations adopting vaccination passes92 and more large employers in the United States implementing vaccine mandates.93

While it now appears unlikely that large countries will reach overall herd immunity (though some areas might), developments in the United Kingdom during the past few months may help illustrate the prospects for Western countries to transition back toward normalcy.94 Having suffered a wave of cases caused by the Delta variant during June and the first few weeks of July, the country delayed plans to ease many public-health restrictions and eventually did so on July 19, though expansive testing and genomic surveillance remain in place. UK case counts may fluctuate and targeted public-health measures may be reinstated, but our scenario analysis suggests that the country’s renewed transition toward normalcy is likely to continue unless a significant new variant emerges.

The United States, Canada, and much of the European Union are now in the throes of a Delta-driven wave of cases.95 While each country’s situation is different, most have again enacted public-health restrictions, thus reversing their transitions toward normalcy. The trajectory of the epidemic remains uncertain, but the United Kingdom’s experience and estimates of total immunity suggest that many of these countries are likely to see new cases peak late in the third quarter or early in the fourth quarter of 2021. As cases decline, our analysis suggests that the United States, Canada, and the European Union could restart the transition toward normalcy as early as the fourth quarter of 2021, provided that the vaccines used in these countries continue to be effective at preventing severe cases of COVID-19. Allowing for the risk of another new variant and the compound societal risk of a high burden of influenza, respiratory syncytial virus, and other winter respiratory diseases, the question for these countries will be whether they manage to arrive at a different epidemiological endpoint, as we discuss next.

Endemic COVID-19 may be a more realistic endpoint than herd immunity

We have previously written about herd immunity as a likely epidemiological endpoint for some countries, but the Delta variant has put this out of reach in the short term. Instead, it is most likely as of now that countries will reach an alternative epidemiological endpoint, where COVID-19 becomes endemic and societies decide—much as they have with respect to influenza and other diseases—that the ongoing burden of disease is low enough that COVID-19 can be managed as a constant threat rather than an exceptional one requiring society-defining interventions. One step toward this endpoint could be shifting the focus of public-health efforts from managing case counts to managing severe illnesses and deaths. Singapore’s government has announced that it will make this shift, and more countries may follow its lead.96

Other authors have compared the burden of COVID-19 with that of other diseases, such as influenza, as a way to understand when endemicity might occur.97 In the United States, COVID-19 hospitalization and mortality rates in June and July were nearing the ten-year average rates for influenza but have since risen. Today, the burden of disease caused by COVID-19 in vaccinated people in the United States is similar to or lower than the average burden of influenza over the last decade, while the risks from COVID-19 to unvaccinated people are significantly higher (Exhibit 2). This comparison should be qualified, insofar as the burden of COVID-19 is dynamic, currently increasing, and uneven geographically. It nevertheless helps illustrate the relative threat posed by the two diseases.


Countries experiencing a Delta-driven wave of cases may be more likely to begin managing COVID-19 as an endemic disease after cases go into decline.98 The United Kingdom appears to be making this shift now (though cases there were increasing as of this writing). For the United States and the European Union, scenario analysis suggests that the shift may begin in the fourth quarter of 2021 and continue into early 2022 (Exhibit 3). As it progresses, countries would likely achieve high levels of protection against hospitalization and death as a result of further vaccination efforts (which may be accelerated by fear of the Delta variant) and natural immunity from prior infection. In addition, boosters, full approval of vaccines (rather than emergency-use authorization), authorization of vaccines for children, and a continuation of the trend toward employer and government mandates and incentives for vaccination are all likely to increase immunity.99


Our scenario modeling suggests that although the resulting level of population immunity may not be high enough to achieve herd protection, it would still protect a substantial portion of the population. Most serious cases of COVID-19 would occur in unvaccinated people. Flare-ups and localized epidemics would happen while COVID-19 is managed as an endemic disease, but scenario modeling suggests that these may have less of an effect on the whole of society than the waves seen to date. Booster vaccinations will be important in maintaining immunity levels over time.100 A new variant that substantially evades existing immunity would remain the biggest overall risk.

Countries have varying prospects for reaching the end of the pandemic

Here, we offer a broader geographic view, comparing the current state as of the time of publishing in countries around the world. Our analysis suggests that countries fall into three general groups (within which national conditions can vary to some extent):

1. High-vaccination countries. These countries, primarily in North America and Western Europe, are the ones discussed above.

2. Case controllers. This group includes countries such as Singapore that have been most successful in limiting mortality associated with COVID-19 to date.101 They have typically maintained tight border restrictions and a strong public-health response to imported cases. Their residents have mostly enjoyed long periods of relative normalcy without public-health restrictions, aside from limits on international travel. Some countries in this group, such as Australia, have recently faced a Delta-driven surge in cases, but in absolute terms the burden of disease remains low relative to other countries. Unless these countries choose to maintain their border restrictions (such as hotel-based quarantine) indefinitely, they might accept the risk of endemic COVID-19 after governments determine that a sufficient portion of the population is vaccinated.102 The pace of vaccine rollout varies among the countries, but in many cases reopening of borders may not begin until 2022, dependent in part on public-health outcomes for countries in other groups.103 The shift from a zero-COVID-19 goal to an endemic, low-burden goal may be challenging for some countries.

3. At-risk countries. Mainly comprising most lower-income and many middle-income countries, this is a group of nations that have not yet gained access to enough vaccine doses to cover a large portion of their populations. Estimates of their overall immunity remain low enough that there is still a risk of significant waves of disease. Recent projections suggest that it is likely to take until late 2022 or early 2023 for these countries to achieve high vaccine coverage.104 The possible time frame for them to manage COVID-19 as an endemic disease is less clear.

Globally and nationally, the epidemiological and public-health situation remains dynamic, and the prospects for each country group are subject to uncertainty. Factors that could influence actual outcomes include:

  • the potential for new variants to emerge (for example, a variant that evades vaccine-mediated immunity to the extent that it frequently causes severe disease in the vaccinated and spreads widely would likely have the most significant effect on any country’s prospects for reaching the end of the pandemic)
  • further evidence of waning natural and vaccine-mediated immunity over time, and challenges with rolling out vaccine boosters quickly enough to maintain immunity
  • further challenges with vaccine manufacturing or global rollout
  • changes in the ways that countries define an acceptable burden of disease (for example, setting different targets for disease burden in vaccinated and unvaccinated populations)

The surge of COVID-19 cases resulting from the spread of the Delta variant and from vaccine hesitancy brought a sudden, tragic end to the transition toward normalcy that some countries had begun to make. But the United Kingdom’s experience indicates that a transition toward normalcy may yet be possible before long, at least in countries where the vaccine rollout is well under way. Their task will be determining what burden of disease is low enough to warrant lifting of public-health restrictions, and how to manage the public-health impacts of endemic COVID-19. In countries where vaccination rates remain low, the prospects for ending the pandemic remain largely tied to the availability and administration of additional doses. Expanding the international vaccine rollout remains essential to achieving a postpandemic sense of normalcy worldwide.

About the authors

Sarun Charumilind is a partner in McKinsey’s Philadelphia office, Matt Craven is a partner in the Silicon Valley office, Jessica Lamb is a partner in the New Jersey office, Adam Sabow is a senior partner in the Chicago office, Shubham Singhal is a senior partner in the Detroit office, and Matt Wilson is a senior partner in the New York office.

The authors wish to thank Xavier Azcue, Marie-Renée B-Lajoie, Andrew Doy, Bruce Jia, and Roxana Pamfil for their contributions to this article.

This article was edited by Josh Rosenfield, an executive editor in the New York office.

March 26, 2021

The fall in COVID-19 cases across much of the world over the past ten weeks signals a new dawn in the fight against the disease. Vaccines are proving effective and rapidly scaling, bending the curve in many geographies. This is a fragile dawn, however, with transmission and deaths still high, unequal access to vaccines, and variants of concern threatening to undo progress to date.

The trajectory of UK and US cases has enabled the beginnings of a transition toward normalcy,105 the first and more important of the pandemic’s two endpoints. We expect this transition to continue in the second quarter of 2021 and will likely see many aspects of social and economic life return to the prepandemic normal, consistent with UK Prime Minister Johnson’s staged reopening plan for the United Kingdom106 and US President Biden’s goal of a normal Independence Day.107 We are more confident in this timeline for the United Kingdom than for the United States, given that the first has already experienced a wave driven by a more infectious variant, whereas the latter could still face one. Parts of the European Union have recently faced setbacks: fewer doses in arms than in the United Kingdom or United States, a new wave of cases, and new lockdowns. A transition toward normalcy is mostly likely in Europe during the late second or third quarter of 2021. The timing will probably vary by country, depending on accelerating vaccine supplies, the impact of vaccinations on hospitalization rates, and the occurrence (or not) of new waves driven by new variants.

Herd immunity, the second endpoint, is most likely in the third quarter for the United Kingdom and the United States and in the fourth quarter for the European Union, with the difference driven by a more limited vaccine availability in the European Union. However, the risks to these timelines are real—herd immunity may not be achieved by the end of the year if vaccine hesitancy is high, if countries experience disruptions in vaccine supply, or if a variant that renders existing vaccines less effective spreads widely. And herd immunity may look different in different parts of the world, ranging from strong nationwide or regional protection to temporary or oscillating immunity to some countries not reaching herd immunity over the medium term.

In this article, we’ll review the developments since our last perspective (January 21), offer an outlook for each of the three geographies, assess risks, and outline what the end of the pandemic might look like.

Recent developments and their impact on timelines

The past month or two have seen seven important developments:

  1. Vaccines work. We have growing evidence that vaccines are effective, as real-world data from Israel and the United Kingdom validate the clinical-trial results by showing a sharp reduction in hospitalizations and deaths.108 Emerging evidence also indicates that vaccines likely reduce transmission considerably, though not to the same degree that they prevent severe disease.109
  2. The vaccine rollout is improving. Massive inoculation programs have accelerated, especially in the United Kingdom. As of March 15, the United Kingdom has administered 39 doses per 100 people in the total population; the corresponding figures for the United States and the European Union are 33 and 12 per 100, respectively.110 Just as important: sentiment about vaccine adoption is improving.111
  3. More vaccines are coming. Johnson & Johnson’s one-shot vaccine appears to be highly effective against severe COVID-19 and received Emergency Use Authorization in the United States on February 27.112 Novavax’s vaccine is now in Phase III trials; preliminary results suggest it was highly effective in the United Kingdom but less so in South Africa.113 All of that makes it increasingly clear that the United Kingdom and the United States will have enough doses to vaccinate all adults by the end of the second quarter, and Europe should achieve the same milestone by the end of the third quarter, assuming no major vaccines are withdrawn. Further, vaccine trials on children aged 12 and up are well underway, and new trials on babies and children six months and older raise the possibility of pediatric vaccination, which would add to the population that could potentially contribute to herd immunity.114
  4. Therapeutics are poised to make more of a difference. A new wave of COVID-19 therapeutics, including those from Eli Lilly,115 Merck–Ridgeback,116 and Vir Biotechnology,117 have produced positive data or received Emergency Use Authorization. The emerging data from these treatments suggest they have the potential to materially reduce hospitalizations and deaths for cases that do occur, accelerating a transition toward normalcy.
  5. New cases and deaths are lower—but still high. New cases, hospitalizations, and deaths have dropped dramatically—by 79 percent and 89 percent, respectively, in the United States and the United Kingdom from the January peak, as of March 15.118 This trajectory has amplified discussions of a transition toward normalcy in both countries. The bend in the curve is fragile, however. Much of the decline in Europe has followed strict lockdowns; but lockdowns and other nonpharmaceutical interventions are still confoundingly difficult to get right, and even now, multiple European countries are experiencing upward case trends. And US deaths are still averaging 1,000 per day, many times higher than average daily flu deaths.119
  6. It is increasingly clear that more infectious variants of concern120 may drive a new wave of cases in the coming months. The United Kingdom is in a relatively favorable position; cases are declining in spite of the high prevalence of the B.1.1.7 variant, suggesting that the country has a demonstrable ability to control the spread of more infectious variants. In contrast, the United States and parts of the European Union appear to have an increasing prevalence of B. The potential for a variant-driven wave of US cases and ongoing spread in Europe in the coming months is real.
  7. Variants may also reduce vaccine efficacy or enable reinfection. Data from the AstraZeneca vaccine trial in South Africa highlight the potential for variants such as B.1.351 and P.1 to reduce the efficacy of vaccines.122 Other vaccine data, including those from Novavax and Johnson & Johnson, show a more modest reduction in efficacy, especially against severe disease caused by these variants.123 There is also early evidence of mutations arising independently in the United States that may reduce the efficacy of vaccines.124 These variants also appear to be more infectious than the original wild-type strain. These initial findings are based on very small sample sizes and may change as more information becomes available; we still do not know the impact of vaccines against severe disease from these strains. But if these results hold up, the spread of strains against which existing vaccines are substantially less effective would be a significant risk to lives and could delay the end of the pandemic.

What’s the net impact of all these developments? The data continue to indicate, as stated in our earlier perspectives, that a significant transition toward UK and US normalcy will occur in the second quarter of 2021, although the potential for a variant-driven wave in the United States is real and would blunt the transition (Exhibit 1). Potential herd-immunity timelines are bifurcating as a result of growth in variants that may reduce vaccine efficacy. If the variants turn out to be a minor factor (they only reduce vaccine efficacy modestly, or they don’t spread widely), then herd immunity in the second half of the year is likely for both countries—and is more likely in the third quarter than the fourth. However, if the impact of these variants is significant, we could see timelines significantly prolonging into late 2021 or beyond.

Earlier peak, longer tail: Q3 now likelier for herd immunity, given vaccine availability, but variants of concern could prolong the end.

How does this vary by geography?

Most of our analysis in this series has focused on the United Kingdom and the United States, which continue to move down a similar path. The end of Europe’s pandemic may come somewhat later, and other countries’ outlooks will depend on several variables.

European Union. Here, as in other regions, the timing of access to vaccines will be the biggest driver of the end of the pandemic. Levels of natural immunity from prior infection vary within the European Union but are generally in the same range as in the United Kingdom and the United States.125 Seasonality is likely to work in similar ways. And public interest in vaccination appears to be similar too, even in countries such as France, where interest in vaccination was significantly lower than in other countries in the region but may now be improving.126 The prevalence of the more infectious B.1.1.7 variant varies by country; most countries with cases are between the high UK levels and lower US levels.

Exhibit 2 lays out the likely timing of vaccine availability in the European Union. Broadly speaking, availability will be similar to that of the United Kingdom and the United States, but EU countries may need to wait a few months longer to vaccinate all adults. Please note that this would be subject to change and further delay if the Oxford–AstraZeneca vaccine remains suspended in multiple countries following concerns about blood clots; WHO has confirmed its continued support of the vaccine.127

Europe’s vaccine supply seems sufficient to vaccinate 68 percent of adults by June 2021.

Given that sufficient vaccine doses are available to vaccinate the highest-risk populations in the coming months, we expect to see the EU transition to normalcy during the second quarter of the year, although the start of this transition may be delayed until late in the quarter by a new wave of cases in some countries. A key difference for the European Union, as compared with the United Kingdom and the United States: herd immunity is more likely in the fourth quarter than the third quarter, given the likely timeline of vaccine delivery (Exhibit 3).

Vaccines should be available for most adults in the European Union, United Kingdom, and United States, but timing varies.

Rest of the world. While the European Union, the United Kingdom, and the United States have had broadly similar COVID-19 experiences, other parts of the world look very different. Countries like New Zealand have avoided significant COVID-19-associated mortality but appear to be further from herd immunity because so few New Zealanders have infection-driven immunity to SARS-CoV-2. On the other hand, if vaccine uptake is fast, New Zealand might achieve a longer-lasting vaccine-based herd immunity. A second factor is seasonality: the timing of seasonality-driven changes will be different in tropical locations and the Southern Hemisphere. A third is demographics: while the younger populations of many lower-income countries have led to lower COVID-19-associated mortality, they also make it harder for adult-only vaccination programs to drive herd immunity. And perhaps most importantly for timelines, access to vaccines is unequal. While COVAX and other access initiatives are working to close the gap, many low-income countries may not receive enough doses to vaccinate all adults until well into 2022.128 The world is on pace to manufacture enough doses for 80 percent of the global population—or close to 100 percent of the adult population—by the end of 2021, but the distribution of these doses may continue to be asymmetric.

Exhibit 4 provides a global view of seven factors that are likely to drive herd-immunity timelines for the rest of the world. These factors include the following:

  • Population vaccinated—the proportion of people who have received the vaccine so far
  • Vaccine courses secured—the additional supplies for which a country has contracted
  • Supply-chain readiness
  • Consumer vaccine sentiment—the public’s willingness to be vaccinated
  • Population under 19 years of age—a greater proportion of children makes a transition toward normalcy easier to achieve but herd immunity more difficult
  • Natural immunity, or the rate of prior COVID-19 infection—higher historical infection rates decrease the vaccination rate needed to achieve herd immunity
  • Prevalence of variants of concern

Risks to herd immunity

Herd immunity requires that enough people be simultaneously immune to SARS-CoV-2 to prevent widespread ongoing transmission. While data indicate that the most likely scenario is to reach this state on the timelines described above, five risks could delay progress.

First, vaccine adoption may prove lower than expected. That could happen if a real or perceived safety issue increases hesitancy or if younger populations see little reason to be vaccinated once older cohorts are protected and a transition toward normalcy is well underway. Second, herd immunity relies on the efficacy of vaccines at reducing transmission (rather than the usually reported efficacy at preventing disease in the vaccinated person). While initial data suggest that COVID-19 vaccines do block significant transmission,129 the efficacy rate may not prove high enough to drive herd immunity. Third, the duration of vaccine-mediated immunity may prove shorter than anticipated, making it hard to reach the necessary threshold for simultaneous immunity. Fourth, supply-chain disruptions and delays are real, and could produce supply shocks and interfere with timelines. Fifth, and most concerning, variants that reduce the efficacy of vaccines or the benefits of natural immunity may spread widely. Some initial data offer concerning evidence that B.1.351 and P.1 may be examples of such variants, although recent Novavax data (with a small sample size) offer some reassurance that its vaccine is effective against severe disease caused by B1.351.130 Similarly, limited data from the Pfizer-BioNTech and AstraZeneca vaccines show evidence of some protection against P.1.131   

These five factors combined mean that there is still a meaningful chance that herd immunity is not reached in the medium term.

From theory to practice: What the ‘end’ might look like

The pandemic’s two endpoints, a transition toward normalcy and herd immunity, may look different in different places. As the name implies, a transition will include a series of steps that will gradually normalize aspects of social and economic life. The order and pace of these steps will vary by geography. Not everyone will immediately resume all of their prepandemic activities; rather, there will be a noticeable shift toward more of them. Steps may include a return to fully in-classroom education, fewer restrictions on the operations of bars and restaurants, more gatherings with larger groups of people, the reopening of offices, and fewer prohibitions on interregional or international travel. The United Kingdom’s plan for reopening provides an example of the stepwise manner in which a transition to normalcy is likely to occur.132

Herd immunity will represent a more definitive end to the pandemic. Isolated cases may still occur—indeed, the virus may continue to circulate for one or more quarters after herd immunity is reached. But with herd immunity, population-wide public-health measures can be phased out. As populations get closer to this state, it may be helpful to introduce some nuance to what we mean by the term.

  1. Nationwide herd immunity. The full population is well protected so that the country experiences, at most, occasional small flare-ups of disease. This scenario is most likely in smaller countries where immunity to COVID-19 can become uniformly high.
  2. Regional herd immunity. Some regions, states, or cities are well protected, while others experience ongoing outbreaks of COVID-19. In large, diverse countries like the United States, this situation is especially easy to imagine.
  3. Temporary herd immunity. A population or region achieves herd immunity for some period, but as variants are introduced, against which prior immunity is less effective, a new wave of cases is launched. Another potential trigger for such a wave could come as immunity (particularly natural immunity) wanes. As the number of new cases of COVID-19 falls globally, the rate of emergence of important variants should also decrease, but some risk will remain.
  4. Endemicity. A region fails to achieve herd immunity. Endemicity is most likely in places where vaccine access is limited, where few people choose to be vaccinated, if the duration of immunity is short, or variants that reduce vaccine efficacy are common and widespread. Endemicity might include cyclic, seasonal waves of disease, broadly similar to the flu, or a multiyear cycle of resurgence.

The next few years are likely to see a combination of some or all of these options around the world. Given the likely timing of herd immunity in various geographies and the uncertain duration of protection from vaccines (both duration of immune response and efficacy versus new variants), it is likely that some measures such as booster vaccines are likely to be required indefinitely. Herd immunity is not the same as eradication. SARS-CoV-2 will continue to exist. Even when a country reaches herd immunity, ongoing surveillance, booster vaccines, and potentially other measures may be needed.

A year ago, the world was coming to terms with a long, difficult journey ahead. Twelve months later, the end of the pandemic is in sight for some parts of the world. It’s much too soon to declare victory, however. We hope that our perspectives prove useful to leaders as they set policy and strategy; we will continue to update the series.

Download the article here.

About the authors

Sarun Charumilind and Jessica Lamb are both partners in McKinsey’s Philadelphia office, Matt Craven is a partner in the Silicon Valley office, Adam Sabow is a senior partner in the Chicago office, and Matt Wilson is a senior partner in the New York office.

The authors wish to thank Xavier Azcue for his contributions to this article.

This article was edited by Mark Staples, an executive editor in the New York office.

January 20, 2021

This article updates our earlier perspectives on when the coronavirus pandemic will end. Transition toward normalcy in the United States remains most likely in the second quarter of 2021 and herd immunity in the third and fourth quarters, but the emergence of new strains and a slow start to vaccine rollout raise real risks to both timelines. We also add a perspective for the United Kingdom.

The past five weeks have brought an array of conflicting news on the COVID-19 pandemic, affecting our estimates about when the coronavirus pandemic will end. Margaret Keenan, a British nonagenarian, made history on December 8 by becoming the first person to receive the Pfizer-BioNTech vaccine for SARS-CoV-2 outside a clinical trial.133 Since then, several other vaccines have been authorized for use around the world. Sufficient doses are likely to be available to vaccinate high-risk populations in the United States in the first half of 2021. In parallel, however, more-infectious strains of the virus have been detected in South Africa, the United Kingdom, and elsewhere and have spread to an increasing number of countries.134 And the initial rollout of vaccines has been slower than hoped in many places.135

While the United States could still achieve herd immunity in the third or fourth quarter of 2021 (in line with the peak probability in our previous estimates), the emergence of more-infectious variants of SARS-CoV-2 increases the risk that this milestone will not be achieved until later. More-infectious viruses require that a higher percentage of people be simultaneously immune to reach herd immunity.136 While a more infectious variant likely means more people are acquiring natural immunity through infection (despite ongoing efforts to minimize new cases), the net impact of more-infectious strains is likely to be that a higher portion of the population needs to be vaccinated, which may take more time.

We still believe that the United States can transition toward normalcy during the second quarter of 2021, but the same risks also threaten this timeline. A transition toward normalcy would be driven by a combination of seasonality aiding a decline in cases and early vaccine doses helping reduce mortality by protecting those at greatest risk of serious illness. As COVID-19’s impact on health wanes, we are likely to see greater normalization of social and economic life. Data on the availability of vaccine doses in the United States increase confidence that this is possible, but the slow start to the vaccine rollout reinforces that success is by no means guaranteed.

This article describes “most likely” timelines for when the coronavirus pandemic will end. It is now harder to imagine the United States or United Kingdom transitioning to normalcy before second quarter 2021 or reaching herd immunity before third quarter 2021. But a number of other factors could delay the timelines beyond those described, including unexpected safety issues emerging with early vaccines, significant manufacturing or supply-chain delays, continued slow adoption, further mutation, or a shorter-than-anticipated duration of vaccine-conferred immunity. Herd immunity will also require vaccines to be effective in reducing transmission of SARS-CoV-2, not just in protecting vaccinated individuals from getting sick. This is likely, but has not yet been proven at scale.137

Tania Adam webinar


A McKinsey Live event on ‘COVID-19 vaccines are here. What does that mean for your organization?’

Herd immunity

More-infectious strains raise the bar

Herd immunity to a pathogen is achieved when a sufficient portion of a population is simultaneously immune to prevent sustained transmission. The threshold to achieve it is governed by a number of factors, including the transmissibility of the disease.138 More-infectious strains of SARS-CoV-2 therefore raise the bar on herd immunity. The virus has been mutating since it was identified a year ago. The concerning development in recent weeks has been the confirmation of new strains in South Africa, the United Kingdom, and elsewhere that combine multiple mutations and have different profiles. While data are still emerging, initial estimates suggest that the transmissibility rate of the UK strain is 40 to 80 percent higher than that of the original SARS-CoV-2 strain, and that transmission rates could be higher among children too.139140 There is no evidence of higher case fatality with either new strain, but there are fears that new strains may affect how antibodies bind to the virus and may reduce the efficacy of vaccines or antibody treatments developed over the past few months. More data are likely to emerge on this in the weeks ahead.

If these strains become dominant, they may cause a material delay in reaching herd immunity. While many people are acquiring natural immunity through infection, variants with enhanced transmissibility, if they predominate among all strains, could increase the proportion of people who need to be simultaneously immune to achieve herd immunity by ten to 20 percentage points, and increase vaccine coverage levels needed to 65 to 80 percent of the population (or 78 to 95 percent of those over 12 years old).141 More detail is shown in Exhibit 1 below.

Vaccinating more people is a nonlinear challenge. Consumer surveys suggest that a portion of the population is cautious about vaccination. Increasing coverage from 70 to 80 percent is therefore harder than increasing from 60 to 70 percent. Because more-transmissible variants raise herd-immunity thresholds, there will also be less tolerance for low vaccine effectiveness. For example, with a variant that is 40 to 80 percent more transmissible, vaccine efficacy of 90 percent would require 83 to 100 percent of those over 12 to be vaccinated; efficacy of anything less than 75 percent would make herd immunity likely unachievable through vaccination of only those over 12.

While the variant of concern appears to be most widespread in the United Kingdom, it has been detected in over 30 countries, many of which (including the United States) have limited capacity for genetic sequencing. As a result, we may be significantly underestimating its spread.142143 The strain is likely to continue spreading in the coming months, propelled by its reproductive advantage over the original. This appears to have occurred in southern England over the past few months. If new strains predominate, they could lead to a longer timeline to herd immunity.

Vaccine rollout: A slow start, but there is still time to improve

The speed of COVID-19 vaccine development has been an unqualified success. The approval, in at least one country, of vaccines made by Pfizer and BioNTech, Moderna, Oxford and AstraZeneca, Sinopharm, Serum Institute, Bharat Biotech, Gamaleya, and others within a year of viral sequencing smashed all records for development timelines. But rollout is off to a slow start. While countries such as Israel have shown what is possible, the United States has fallen behind its targets.144145 It is still early days, and there is time to accelerate, but there is little margin for error if the United States is to achieve herd immunity in third quarter 2021. In addition, not all regions are adhering closely to manufacturer dosing protocols—for example, delaying second doses or giving a first dose from one manufacturer followed by a second from another—and the impact of that is unclear. These approaches could reduce mortality in the short term by broadening access, but they could also delay herd immunity if, for example, a delayed second dose reduces efficacy. It’s also possible that once most people in the highest-risk groups have received vaccinations, the pace of vaccination will slow if lower-risk groups do not embrace the opportunity.

We believe that herd immunity in the United States is still most likely in third or fourth quarter 2021, but that the chance of delay until first quarter 2022 or beyond has increased (Exhibit 2). There is relatively little chance of achieving herd immunity before then. Even later herd immunity remains possible if other challenges arise, especially vaccine safety concerns or ambivalence to vaccination following a transition toward normalcy. This potential delay represents a call to action for policy makers, both in terms of the pace of the vaccine rollout and how new strains are managed.

Probability of herd immunity to COVID-19 for the United Kingdom and United States, by date; factors affecting the date (chart)

More-transmissible variants and slower initial vaccine rollout increase downside risk to the Q3–Q4 2021 timeline to achieve herd immunity.

Chart: Probability of herd immunity1 to COVID-19 pandemic for the United Kingdom and United States2 by quarter (illustrative)

Two skewed bell curves estimate when herd immunity will occur.

  • 11/23/20 estimate. From Q4 2020 to Q1 2021 the probability of herd immunity is negligible. The curve rises sharply from Q1 2021 to a peak between Q3 and Q4 2021. The long tail of the curve shows falling probabilities to Q3 2023 and beyond.
  • 1/15/21 estimate. The start, peak, and tail of this curve are almost the same as the previous one. A flatter curve indicates herd immunity will probably occur one month later.

Early herd immunity if:

  • Vaccine rollout and adoption are faster than expected
  • Natural immunity is significantly higher than realized
  • More-transmissible variants lead to higher rates of natural immunity

Peak probability of herd immunity driven by:

  • US Biologics License Applications (BLA) with full approval by March/April 2021 or earlier
  • Approximately 3–9 months for manufacturing, distribution, and sufficient adoption to reach herd immunity

Later herd immunity if one or more of the following occur:

  • Safety issues delay BLA
  • Manufacturing/supply-chain issues slow rollout
  • More-infectious variants raise the threshold for achieving herd immunity
  • Adoption is slower than anticipated
  • Duration of immunity is short
  • Vaccine prevents disease progression but does not meaningfully reduce transmission


1Herd immunity is achieved when a sufficient portion of a population is simultaneously immune to prevent sustained transmission. At this point, significant, ongoing public-health measures are not needed to prevent future spikes in disease and mortality (this might be achieved while there are still a number of people in particular communities who still have the disease, as is the case with measles).

2Timeline to functional end is likely to vary somewhat based on geography.

McKinsey & Company

Increasing recognition of medium-term endemicity

While many parts of the world are expected to reach herd immunity against COVID-19, there is increasing consensus that globally, SARS-CoV-2 is likely to remain endemic in the medium term. David Heymann, the chairman of the World Health Organization’s Strategic and Technical Advisory Group for Infectious Hazards, noted in December that endemicity may be the “destiny” of this virus.146 This might make COVID-19 analogous to measles—a disease that causes intermittent, limited outbreaks in countries with well-developed vaccination programs but significant ongoing disease in parts of the world where access to vaccines is more limited. It is also possible that COVID-19 will be seasonal, with predictable annual peaks in parts of the world where it is endemic.147

Transition toward normalcy

A transition toward normalcy will occur when COVID-19 mortality falls and the disease is de-exceptionalized in society. COVID-19 will not disappear during this transition, but will become a more normal part of the baseline disease burden in society (like flu, for example), rather than a special threat requiring exceptional societal response. During this transition, controlling the spread of SARS-CoV-2 will still require public-health measures (such as continued COVID-19 testing and mask use in many settings), but mortality will fall significantly, allowing greater normalization of business and social activities. This will be driven by a combination of early vaccine rollout (which, being directed first at those at greatest risk, should reduce deaths faster than cases), seasonality, increasing natural immunity, and stronger public-health response.

Increasing clarity on the availability of vaccine doses during the first half of 2021 in the United States improves the odds of an early transition toward normalcy. As Exhibit 3 shows, Pfizer and Moderna are expected to deliver sufficient vaccine doses to vaccinate all high-risk Americans during the first half of the year. This does not account for other vaccines that are likely to become available, including those approved in other markets (for example, Oxford-AstraZeneca) or others that are likely to report clinical trial data in the first quarter of 2021 (including Johnson & Johnson and Novavax). Older people are generally more willing to be vaccinated than the general population. However, slow initial rollout of the vaccines and the spread of more infectious variants increase the risk that significant mortality continues in the second quarter, blunting a transition to normalcy.

In the United States, doses committed by Pfizer and Moderna by July 31, 2021 are approximately enough for population in phases 1a–c (chart)

In the United States, doses committed by Pfizer and Moderna by July 31 are approximately enough for population in phases 1a–c.

Estimate of supply needed, doses1 for US population in phases 1 and 24

  • 48 million for phase 1a
  • 98 million for phase 1b
  • 56 million for phase 1c: ages 65–74 (through 3/31/21)
  • 202 million for phase 1c: other (4/1/21–7/31/21)
  • ~160 million for phase 2 (2021 Q3/4)
  • Total ~560 million

Estimate of supply available

Delivery deadlines for vaccines with Emergency Use Authorization in the United States, millions of doses1 (illustrative)
Company By 3/31/21 Between 4/1/21 and 6/30/21 Between 7/1/21 and 7/31/21 Total by 7/31/21 2021 Q3/4
Pfizer 100 70 30 200
Moderna 100 100 200
Total of Pfizer and Moderna 200 170 30 400
Potential AstraZeneca, J&J, and Novavax doses2 unspecified additional number
Options to purchase3 300 Pfizer, 400 Moderna


1Two doses needed per person.

2Subject to regulatory authorization.

3Timing not specified.

4According to CDC ACIP interim recommendations (December 22, 2020), will vary as individual states are making their own decisions (CDC phase 1a = healthcare personnel, long-term care facility residents; CDC phase 1b = frontline essential workers, persons aged ≥75 years; CDC phase 1c = persons aged 65–74 years; persons aged 16–64 years with high-risk medical conditions; essential workers not recommended for vaccination in phase 1b); phase 2 estimate based on 2019 census population estimate of persons aged ≥16, less population accounted for in CDC estimates of persons covered in phases 1a–c; CDC and Operation Warp Speed vaccination guidelines may evolve over time.

Source: Bloomberg; DC; HHS; Moderna; Pfizer; Reuters; WSJ

McKinsey & Company

Taking these variables into account, we still believe that a transition toward normalcy is likely during second quarter for the United States, but that downside risks have increased. If early vaccine doses reach a significant percentage of high-risk elderly individuals by the end of quarter one, the combination of protecting these groups and the arrival of spring in the northern hemisphere should improve the situation compared with where the United States is now. Depending on vaccination progress over the summer (whether the United States is on the earlier or later end of the herd immunity window), there may be a smaller fall wave of disease in third to fourth quarter 2021.

While the potential for a transition toward normalcy in just a few months is encouraging, many signs suggest that the next six to eight weeks will be difficult. Case and death numbers are at or near all-time highs in many locations, new variants may accelerate short-term transmission, and vaccine rollout has not yet proceeded far enough to protect much of the population. Strong public-health measures will remain critical to saving lives during this period.

Timeline for the United Kingdom

We see similar dynamics in the United Kingdom. Three factors lead us to believe that timelines for herd immunity and transition toward normalcy in the United Kingdom will be similar to those in the United States. First, access to vaccines is sufficient to immunize a large percentage of both the US and UK populations during 2021. Second, public willingness to be vaccinated is generally similar between the two countries.148 Third, the fraction of US and UK residents who already have natural immunity from prior infection is in the same range (with significant variability among regions within countries).149

The variant of concern represents a potential source of difference. While it is known to be highly prevalent in the United Kingdom and present in the United States, there is a significant chance that it will predominate throughout the United States over the coming months. All else being equal, countries with a higher proportion of more-infectious variants—assuming they increase public-health measures to handle them—are likely to achieve herd immunity later.

We will add a perspective for other parts of the world, including the rest of Western Europe, in future updates to this article.

Twelve months ago, most people weren’t thinking about COVID-19. Today, much of the world is intensely focused on it, but we can reasonably expect the imminent threat to abate. Much work remains to be done. In the short term, public-health measures can help control the pandemic, but even when herd immunity is achieved, managing the risk of COVID-19 will require monitoring, potential revaccination, and treatment of isolated cases. Every country has its own COVID-19 story, but those stories will eventually reach some kind of ending.

Download the article here.

About the authors

Sarun Charumilind and Jessica Lamb are both partners in McKinsey’s Philadelphia office, Matt Craven is a partner in the Silicon Valley office, Adam Sabow is a senior partner in the Chicago office, and Matt Wilson is a senior partner in the New York office.

The authors wish to thank Xavier Azcue, Brian Hencke, David Meredith, Michalis Michaelides, Anthony Ramirez, Virginia Simmons, Konstantinos Tsakalis, and Lieven Van der Veken for their contributions to this article.

November 23, 2020

Our November 23 update takes on the questions raised by recent news: When will vaccines be available? And is the end of COVID-19 nearer?

Since we published our first outlook, on September 21st, the COVID-19 pandemic has raged on, with more than 25 million additional cases and more than 400,000 additional deaths. While the situation looks somewhat better in parts of the Southern Hemisphere, much of Europe and North America is in the midst of a “fall wave,” with the prospect of a difficult winter ahead. Yet the past two weeks have brought renewed hope, headlined by final data from the Pfizer/BioNTech150 vaccine trial and interim data from the Moderna trial, both showing efficacy of approximately 95 percent151; and progress on therapeutics. Is an earlier end to the pandemic now more likely?

The short answer is that the latest developments serve mainly to reduce the uncertainty of the timeline (Exhibit 1). The positive readouts from the vaccine trials mean that the United States will most likely reach an epidemiological end to the pandemic (herd immunity) in Q3 or Q4 2021. An earlier timeline to reach herd immunity—for example, Q1/Q2 of 2021—is now less likely, as is a later timeline (2022). If we are able to pair these vaccines with more effective implementation of public-health measures and effective scale-up of new treatments and diagnostics, alongside the benefits of seasonality, we may also be able to reduce mortality enough in Q2 to enable the United States to transition toward normalcy. (See sidebar “Two endpoints” for our definitions.)

Main effect of recent news is to increase confidence in Q3–Q4 2021 as most likely timeline to achieve herd immunity.

A secondary effect of the recent vaccine trials is to make Q3 2021 more likely for herd immunity than Q4. That said, major questions are still outstanding, even about vaccines, such as long-term safety, timely and effective distribution, and vaccine acceptance by the population, to say nothing of lingering epidemiological questions such as the duration of immunity.

These are estimates for the United States, which is likely to have fast and ready access to vaccines. We will consider timelines for other countries in forthcoming updates; they will vary based on the timing of access and distribution of vaccines and other factors. In this update, we review the most recent findings, look deeper at five implications of the ongoing scientific research, and discuss why our timeline estimates have not shifted meaningfully.

Revelations from vaccine and antibody trials

The world has cheered announcements over the past two weeks by Pfizer and its partner BioNTech, and from Moderna. Their COVID-19 vaccine candidates are showing efficacy rates that are higher than many dared hope for. One is a final result, and the other is an initial result whose sample size is large enough to give reasonable confidence in the data. At about 95 percent, efficacy is higher than expected by most experts.152 It exceeds the optimistic case that we included in our September article. Higher efficacy provides greater benefit to any vaccinated individual and may help to encourage uptake among some segments of the population. It also reduces the fraction of the population required to reach herd immunity. Moderna also announced that its vaccine is more shelf-stable than expected and would need only refrigeration to keep it stable for 30 days—another piece of good news. Finally, there are a number of other vaccines in late-stage trials from which data is expected in the coming months.

Caution is still warranted. The safety records of the Pfizer and Moderna vaccines appear promising so far (no serious side effects reported), but the coming months will provide a fuller picture as the sample size grows. We don’t yet know how long the protection the vaccines offer will last. The Pfizer trial has enrolled some children (ages 12 and older), but efficacy in those under 18 remains unclear.

Beyond vaccines, science is also progressing in therapeutics for COVID-19. For example, Eli Lilly’s antibody bamlanivimab was granted Emergency Use Authorization (EUA) by the US Food and Drug Administration on November 9,153 and Regeneron’s EUA for its antibody cocktail REGN-COV2 for EUA was approved on November 22. Emerging data on these antibodies suggest that they can reduce the need for hospitalization of high-risk patients, and hold potential for post-exposure prophylaxis.154 While they are not recommended for use in hospitalized patients, these antibodies add to the growing armamentarium of treatments and protocols for COVID-19, where every incremental advance could help to reduce mortality. Collectively, these treatments and changes in clinical practice have lowered mortality for those hospitalized by 18 percent or more.155

Looking deep into the data

Research and findings of the past two months have shed light on a number of uncertainties and in some cases have raised new questions. Here we review five implications; each has helped refine our probability estimates for the COVID-19 pandemic timeline.

Vaccine age restrictions elevate coverage requirements to reach herd immunity

It appears that the two vaccines mentioned will be indicated first for use in adults.156 It’s not clear when use in children will be indicated. One consequence is that the vaccines’ contribution to population-wide herd immunity will depend on adults, at least until vaccines are approved for use in younger populations. If vaccines are efficacious, safe, and distributed to all ages, vaccine coverage rates of about 45 to 65 percent—in combination with projected levels of natural immunity—could achieve herd immunity (Exhibit 2).

On the other hand, if vaccines are efficacious but distributed only to adults, who comprise only 76 percent of the US population,157 then higher vaccine coverage rates—approximately 60 to 85 percent—could be required to achieve herd immunity.

Another consequence is that older children, who have twice the COVID-19 incidence of younger children and who have higher viral loads (and therefore greater potential contagiousness) than adults158 may not have immediate access to vaccines.

We recognize that calculating herd immunity thresholds is complex. Basic formulas fail to account for variations in the way populations interact in different places.159 For this reason we include relatively wide ranges.

Unclear impact of vaccines on transmission could raise the bar on coverage

Vaccine trials and regulatory approval will be based on safety and efficacy in reducing virologically confirmed, symptomatic disease among individuals.160 That’s not the same as reducing transmission. This distinction will have much to say about whether the United States reaches normalcy in Q2 or Q3 of 2021. In practice, we have data on whether people who are vaccinated are less likely to get sick with COVID-19 (and less likely to get severe disease), but we won’t have data on how likely they are to transmit to others. It’s an important distinction because what will drive herd immunity is reduction in transmission. If vaccines are only 75 percent effective at reducing transmission, then coverage of about 60 to 80 percent of the population will be needed for herd immunity. And if a vaccine is only 50 percent effective at reducing transmission, coverage of over 90 percent would be required (Exhibit 3).

Leaders should be alert to possible scenarios of lower vaccine efficacy.

Wide variations in local seroprevalence suggest heterogeneous paths to herd immunity

Improved estimates of seroprevalence are increasingly available for many regions. They vary widely, from as low as 1 to 2 percent in some states like Colorado and Kansas to 14 to 20 percent in New Jersey and New York.161 Because achieving herd immunity relies in part on a population’s natural immunity, it appears that some locations are closer to herd immunity than others (and have likely also experienced worse impact on public health to date.) Based on a range of likely vaccine scenarios and the fact that those with prior exposure to SARS-CoV-2 will still be eligible for vaccination, every ten percentage-point increase in seroprevalence could roughly translate into a one-month acceleration of the timeline to the epidemiological endpoint.

However, it is possible that areas with higher seroprevalence may also have higher thresholds for herd immunity, because their populations may mix more,162 which could have contributed to higher seroprevalence to begin with. If that’s true, then while they are further along, they may also have further to go. Well-executed distribution of effective vaccines will still be paramount.

Potentially shorter duration of immunity could prolong the path to the ‘end’

Earlier in the pandemic, it was unclear how long immunity after COVID-19 infection would last. Duration of immunity matters, obviously; for instance, our modeling suggests that if natural immunity to COVID-19 lasts six to nine months, as opposed to multiple years (like tetanus) or lifelong (like measles), herd immunity is unlikely to be achieved unless adult vaccination rates approach 85 percent. While COVID-19 reinfection is documented but rare, there are now population-level studies that question the durability of immunity. Antibody levels may wane after just two months, according to some studies, while a United Kingdom population-monitoring effort reported that antibody prevalence fell by 26 percent over three months.163 The relationship between waning antibodies and reinfection risk remains unclear. Other research suggests that even with waning levels of COVID-19 antibodies, the immune system may still be able to mount a response through other specific B-cell and T-cell immune pathways, where emerging evidence shows much greater durability after six months.164

Manufacturing and supply issues are clearer, but have not vanished

If the initial efficacy data from the Pfizer and Moderna vaccine trials hold up, and if no significant safety issues emerge, then initial demand is likely to be high. Two promising candidates are better than one, but supplies will undoubtedly be constrained in the months following EUA and approval. The situation may be dynamic as vaccines are approved at different times, each with its own considerations in manufacturing and distribution. For example, current data suggest that Moderna’s vaccine is stable at refrigerated temperatures (2 to 8 degrees Celsius) for 30 days and six months at –20 degrees Celsius. Pfizer’s vaccine can be stored in conventional freezers for up to five days, or in its custom shipping coolers for up to 15 days with appropriate handling. Longer-term storage requires freezing at –70 degrees Celsius, requiring special equipment.165 Both Pfizer’s and Moderna’s would be two-dose vaccines, necessitating rigorous follow-up for series completion. These and other complexities create risk of delay. Timelines to reach the desired coverage threshold will be affected by health systems’ abilities to adapt to changing needs and updated information.

The pandemic’s end is more certain, and may be a little nearer

Given all of these variables, where do we net out?

While the winter of 2020/2021 in the Northern Hemisphere will be challenging, we are likely to see mortality rates fall in Q2 (or possibly late Q1) of 2021. Seasonality and associated changes in behavior will begin to work again in our favor in the spring, and the combination of early doses of vaccines targeted to those at highest risk (and the benefits of the Pfizer and Moderna vaccines in reducing severe disease), advances in treatment, expanded use of diagnostics, and better implementation of public-health measures should serve to significantly reduce deaths from COVID-19 in the second quarter. At this stage, when monthly mortality from COVID-19 may start to resemble that of flu in an average year, we may see a transition toward normalcy, albeit with public-health measures still in place.

We are as excited as others about the stunning developments in vaccines. We think Q3 or Q4 of 2021 are even more likely to see herd immunity in the United States. This is based on EUA of one or more high-efficacy vaccines in December 2020 or January 2021, as manufacturers are targeting166; distribution to people at highest risk (healthcare workers, the elderly, and those with comorbidities) in the early months of 2021167;full approval of a vaccine in March or April; and then widespread rollout. Our estimates of three to eight months for manufacturing, distribution, and adoption of sufficient vaccine doses to achieve herd immunity remain unchanged, and suggest that the milestone may be reached between July and December 2021.

Recent developments suggest that herd immunity is less likely to come in early 2021, given that vaccines are arriving roughly on the expected timeline; and the downside scenario stretching into 2022 is also less likely, since efficacy is clearer. The new vaccines may slightly accelerate the timeline—the ongoing surge in cases will likely continue into winter, which would increase natural immunity levels going into Q2. Further, higher-than-expected efficacy may help offset coverage challenges that surveys have suggested. Those two factors could advance the timeline, and make Q3 a little more likely than Q4.

Our estimate is based on the widest possible reading of the current scientific literature and our discussions with public-health experts in the United States and around the world. It’s possible that unforeseen developments such as significantly more infections than expected this winter could lead to earlier herd immunity. And real downside risk remains, especially with respect to duration of immunity and long-term vaccine safety (given the limited data available so far). Herd immunity might not be reached until 2022 or beyond.

Even when herd immunity is achieved, ongoing monitoring, potential revaccination, and treatment of isolated cases will still be needed to control the risk of COVID-19. But these would fall into the category of “normal” infectious disease management—not the society-altering interventions we have all lived through this year. The short term will be hard, but we can reasonably hope for an end to the pandemic in 2021.

Download the article here.

About the authors

Sarun Charumilind and Jessica Lamb are both partners in McKinsey’s Philadelphia office, Matt Craven is a partner in the Silicon Valley office, Adam Sabow is a senior partner in the Chicago office, and Matt Wilson is a senior partner in the New York office.

The authors wish to thank Gaurav Agrawal, Xavier Azcue, Jennifer Heller, Anthony Ramirez, Shubham Singhal, and Rodney Zemmel for their contributions to this article.

This article was edited by Mark Staples, an executive editor in the New York office.

September 21, 2020

Normalcy by spring, and herd immunity by fall? We assess the prospects for an end in 2021.

In 1920, a world wearied by the First World War and sickened by the 1918 flu pandemic desperately sought to move past the struggles and tragedies and start to rebuild lives. People were in search of a “return to normalcy,” as Warren G. Harding put it. Today, nearly every country finds itself in a similar position.

More than eight months and 900,000 deaths into the COVID-19 pandemic,168 people around the world are longing for an end. In our view, there are two important definitions of “end,” each with a separate timeline:

  • An epidemiological end point when herd immunity is achieved. One end point will occur when the proportion of society immune to COVID-19 is sufficient to prevent widespread, ongoing transmission. Many countries are hoping that a vaccine will do the bulk of the work needed to achieve herd immunity. When this end point is reached, the public-health-emergency interventions deployed in 2020 will no longer be needed. While regular revaccinations may be needed, perhaps similar to annual flu shots, the threat of widespread transmission will be gone.
  • A transition to a form of normalcy. A second (and likely, earlier) end point will occur when almost all aspects of social and economic life can resume without fear of ongoing mortality (when a mortality rate is no longer higher than a country’s historical average) or long-term health consequences related to COVID-19. The process will be enabled by tools such as vaccination of the highest-risk populations; rapid, accurate testing; improved therapeutics; and continued strengthening of public-health responses. The next normal won’t look exactly like the old—it might be different in surprising ways, with unexpected contours, and getting there will be gradual—but the transition will enable many familiar scenes, such as air travel, bustling shops, humming factories, full restaurants, and gyms operating at capacity, to resume.

The two ends are related, of course, but not linearly. At the latest, the transition to normal will come when herd immunity is reached. But in regions with strong public-health responses, normalcy can likely come significantly before the epidemiological end of the pandemic.

The timeline to achieve the ends will vary by location. In this article, we’ll explain the criteria that will be key factors in determining when each is reached. In the United States and most other developed economies, the epidemiological end point is most likely to be achieved in the third or fourth quarter of 2021, with the potential to transition to normalcy sooner, possibly in the first or second quarter of 2021. Every day matters. Beyond the impatience that most feel to resume normal life, the longer it takes to remove the constraints on our economies, the greater will be the economic damage.

The epidemiological end point

Most countries have deferred the hope of achieving herd immunity until the arrival of a vaccine. When herd immunity is reached, ongoing public-health interventions for COVID-19 can stop without fear of resurgence. The timing of the end point will vary by country and will be affected by a number of factors:

Consider the first and most crucial variables: the arrival of vaccines, their efficacy, and their adoption. We see four plausible scenarios for vaccine efficacy and adoption, illustrated in Exhibit 1.173 Different combinations of those two factors will drive varying levels of conferred immunity, implying the extent of natural immunity that will be required to reach herd immunity under each scenario. Combinations of efficacy and adoption beyond those shown are possible.

The other variables will also have much to say about the timeline to reach herd immunity (see sidebar, “Key factors affecting the timeline to herd immunity”).

Based on our reading of the current state of the variables and their likely progress in the coming months, we estimate that the most likely time for the United States to achieve herd immunity is the third or fourth quarter of 2021. As we wrote in July 2020, one or more vaccines may receive US Food and Drug Administration Emergency Use Authorization before the end of 2020 (or early in 2021) and the granting of a Biologics License Application (also known as approval) during the first quarter of 2021.

Vaccine distribution to a sufficient portion of a population to induce herd immunity could take place in as few as six months. That will call for rapid availability of hundreds of millions of doses, functioning vaccine supply chains, and peoples’ willingness to be vaccinated during the first half of 2021. We believe that those are all reasonable expectations, based on public statements from vaccine manufacturers and the results of surveys on consumer sentiment about vaccines.174

Herd immunity could be reached as soon as the second quarter of 2021 if vaccines are highly effective and launched smoothly or if significant cross-immunity is discovered in a population (Exhibit 2). (For more on the potential for a faster resolution of the COVID-19 crisis in the United States, see “An optimistic scenario for the US response to COVID-19”.) On the other hand, the epidemiological end of the pandemic might not be reached until 2022 or later if the early vaccine candidates have efficacy or safety issues—or if their distribution and adoption are slow. At worst, we see a long-tail possibility that the United States could be still battling COVID-19 into 2023 and beyond if a constellation of factors (such as low efficacy of vaccines and a short duration of natural immunity) align against us.


The paths to herd immunity in other high-income countries are likely to be broadly similar to the one in the United States. The timelines will vary based on differences in vaccine access and rollout and in levels of natural immunity—and potentially, in levels of cross-immunity and previous coverage of other vaccines, such as the BCG vaccine. Even as some locations reach herd immunity, pockets of endemic COVID-19 disease are likely to remain around the world, for example in areas affected by war or in communities with persistently low adoption of vaccines. In such places, until herd immunity is reached, COVID-19 might be analogous to measles—not a day-to-day threat to most people, but a persistent risk. If immunity wanes—for example, if booster vaccines are not fully adopted—then COVID-19 could become more widely endemic.

The arrival of herd immunity won’t mean a complete end to all public-health interventions. It’s possible that regular revaccinations would be required to maintain immunity, and ongoing surveillance for COVID-19 will be required. But herd immunity would mean that the emergency measures currently in place in many countries could be lifted.

The pace at which governments relax public-health measures will be critical. Some of those measures (such as full lockdowns and restrictions on certain industries) have significant social and economic consequences, and others (such as testing and tracing), while expensive, don’t. Many governments are employing packages of measures that aim to minimize the number of COVID-19 cases and excess mortality while maximizing social and economic degrees of freedom.

The transition to normal

The second end point of the pandemic may be reached earlier than the first. We estimate that the mostly likely time for this to occur is the first or second quarter of 2021 in the United States and other advanced economies. The key factor is diminished mortality.

Society has grown used to tracking the number of COVID-19 infections (the case count). But case counts matter primarily because people are dying from the disease and because those who survive it may suffer long-term health consequences after infection. The latter is an area of scientific uncertainty, but there is concern that some recovered patients will face long-term effects.175

Most countries have made significant progress in reducing the numbers of deaths and hospitalizations associated with COVID-19. Some are close to eliminating excess mortality. Those results have generally been achieved through a combination of moderately effective interventions rather than a single “big bang” (Exhibit 3).

Reduced COVID-19-related mortality might be achieved through many factors simultaneously.

A transition to the next normal, in whatever form that takes, will come gradually when people have confidence that they can do what they used to do without endangering themselves or others. Gaining that confidence will require a continuation of the progress made to reduce mortality and complications, as well as further scientific study regarding long-term health consequences for recovered patients. When confidence is restored, people will again fill bars, restaurants, theaters, and sports venues to full capacity; fly overseas (except for the highest-risk populations); and receive routine medical care at levels similar to those seen prior to the pandemic.

The timing of such a transition will depend on the progress toward herd immunity, as previously detailed (since more people with immunity means fewer deaths and long-term health consequences), and on the effectiveness of a country’s public-health response. Transitions will be gradual. They have already begun in some locations and could be well advanced in most countries by the first or second quarter of 2021. Given the interconnectedness of the global economy, country timelines to normalcy are not fully independent of one another.

To achieve that, we will need to see significant progress on the epidemiological end point, including an effective vaccine receiving Emergency Use Authorization approval during the fourth quarter of 2020 or the first quarter of 2021, followed by a smooth rollout and adoption by high-risk populations. Favorable findings on natural and cross-immunity would help accelerate timelines. Five additional criteria will also contribute to the transition to a form of normalcy—the more of these that are achieved, the faster the milestone is likely to be reached:

  • continued improvement by governments in the application of public-health interventions (such as test and trace) that don’t significantly limit economic and social activities
  • compliance with public-health measures until we achieve herd immunity
  • accurate, widely available, rapid testing that effectively enables specific activities
  • continued advancements in therapeutics (including pre- and postexposure prophylactics) for and clinical management of COVID-19, leading to lower infection-fatality ratios—substantial progress has already been made through a combination of effective drugs, such as dexamethasone and remdesivir, and changes in clinical management
  • public confidence that there aren’t significant long-term health consequences for those who recover from COVID-19

Both the epidemiological and normalcy ends to the COVID-19 pandemic are important. The transition to the next normal will mark an important social and economic milestone, and herd immunity will be a more definitive end to the pandemic. In the United States, while the transition to normal might be accomplished sooner, the epidemiological end point looks most likely to be reached in the second half of 2021. Other advanced economies are probably on similar timetables.

Download the article here.

Explore a career with us