April 2, 2015

7 key changes in the global religious landscape

What will the world’s religious landscape look like a few decades from now? A new Pew Research Center study attempts to answer that question by projecting the changing size of eight major global religious groups through the year 2050 based on a variety of demographic factors.

The study uses data from 198 countries and territories on fertility, age composition and life expectancy. It also looks at rates of religious switching – where data is available – and migration between countries, and puts all of these factors together to provide the best estimates for the future.

There are many storylines in this data, which can be explored through the full report or on our interactive Global Religious Futures website. Here are a few of the key findings:

1Muslims are the fastest-growing major religious group, largely because they have the highest fertility rate and the youngest population. As a result, the Muslim population is expected to increase from 1.6 billion people (23% of the world’s population as of 2010) to 2.76 billion people (30% of all people in 2050). At mid-century, Muslims will nearly equal Christians – the world’s largest religious group – in size.

Christian and Muslim Population Projections

2The share of the world’s population that is Christian is expected to remain steady (at about 31%), but the regional distribution of Christians is forecast to change significantly. Nearly four-in-ten Christians (38%) are projected to live in sub-Saharan Africa in 2050, an increase from the 24% who lived there in 2010. And the percentage of the world’s Christians living in Europe – which fell from 66% in 1910 to 26% in 2010 – will continue to decline, to roughly 16% in 2050.

3The number of religiously unaffiliated people, also known as religious “nones,” is increasing in places such as the United States and Europe, and we project continued growth. Globally, however, the opposite is true: The unaffiliated are expected to decrease as a share of the world’s population between 2010 and 2050 (from 16% to 13%). This is attributable mostly to the relatively old age and low fertility rates of large populations of religious “nones” in Asian countries, particularly China and Japan.

Size of Religious Groups, 2010-2050

4In the United States, Christians will decline from more than three-quarters of the population in 2010 to two-thirds in 2050, with corresponding rises of religious “nones” as well as Muslims, Hindus and others. At mid-century, Judaism will no longer be the largest non-Christian religion in the U.S.: Muslims are projected to be more numerous than people who identify as Jewish on the basis of religion.

5Buddhists, concentrated in Asia, are expected to have a stable population (of just under 500 million) while other religious groups are projected to grow. As a result, Buddhists will decline as a share of the world’s population (from 7% in 2010 to 5% 2050).

6Indonesia is currently home to the world’s largest Muslim population, but that is expected to change. By 2050, the study projects India to be the country with the largest number of Muslims – more than 310 million – even though Hindus will continue to make up a solid majority of India’s population (77%), while Muslims remain a minority (18%). Indonesia will have the third-largest number of Muslims, with Pakistan ranking second.

7The farther into the future we look, the more uncertainty exists, which is why the projections stop at 2050. But if they are extended into the second half of this century, the projections forecast Muslims and Christians to be roughly equal in number around 2070, with Muslims the slightly larger group after that year.

Topics: Demographics, Religious Affiliation, Christians and Christianity, Muslims and Islam, Population Projections, Religiously Unaffiliated, Buddhists and Buddhism, Hindus and Hinduism

  1. Photo of Michael Lipka

    is a senior editor focusing on religion at Pew Research Center.


  1. Anonymous1 year ago

    Great research, thank you. Are these studies tracking religious participation, as well as religious affiliation? That is to say, at least in Europe and America, there has been talk about how many may identify nominally as Christian but attend church only infrequently. Similarly, some in Muslim populations in many countries Europe and Asia say the same about younger generations – they may identify as Muslim as opposed to anything else, but have limited participation at prayers, or do not take part in Ramadan. Can you say whether this is a global trend or just anecdote?

    1. Bob Janis Dillon1 year ago

      Basically, the answer is no: “Some social theorists have suggested that as countries develop economically, more of their inhabitants will move away from religious affiliation. While that has been the general experience in some parts of the world, notably Europe, it is not yet clear whether it is a universal pattern In any case, the projections in this report are not based on theories about economic development leading to secularisation.” The study tracks “religious switching” – i.e. if a Christian later declares themselves a Mulsim or an atheist – but does not track the much more common phenomenon of a decline of participation. So the data is as yet very incomplete. It will be interesting to see Pew do more research like this one: pewresearch.org/fact-tank/2016/0…

  2. Anonymous1 year ago

    This is just fascinating! Thank you for crunching the numbers.

    I do have one question, since you are responding. I hear of anecdotal evidence of declining birth rates in Muslim countries as well. Did you factor this in, or just extrapolate the current birth rates? Thanks.

    1. Michael Lipka1 year ago

      The short answer is yes, we factored in expected fertility declines in countries that have high fertility levels. Here are some links with more details:




      Thanks for your comment,
      Michael Lipka

  3. Mike Daum2 years ago

    Judging by the comments to your article, Pew should next consider measuring religious bigotry and intolerance.

    1. Jerry2 years ago

      Actually Mike they already do this. Last year there was a major study done on similar topics. What would make the work easier Mike if you are truly concern is if ever you are subject to share your thoughts don’t hesitate.

  4. Robert Muscat2 years ago

    There is an important factor that will affect future religion demographics that should have been mentioned in this piece even though it is numerically completely unpredictable: the extent of the future spread of skepticism/atheism. Huge numbers of the relatively young will be increasingly exposed to science (including likely advances in fundamentals of life sciences, and of astronomy). The impact of non-religious thinking has been gathering for a few centuries and is likely to continue to grow. This might not show up in future surveys as bumps in the non-affiliated; large numbers of people who consider their inherited religion an aspect of their “identity,” even today, should only be counted as nominally belonging. The important question is: what do they, will they, think and believe about the “truth-claims” of this identity?

    1. Collin Merenoff2 years ago

      It’s already established fact that life evolved and is still evolving, and that the cosmos is over 13 billion years old and immeasurably vast. There’s nothing for youngsters to debate on these subjects, except perhaps how to silence their contrarian elders. 🙂

      I think the impact of science on the next generation will be in physics, as the evidence against quantum idealism continues to accumulate.

  5. Ravish Khare2 years ago

    This analysis appears flawed. The unaffiliated or the non-religious population is not just due to parents non-religious affiliation, but mainly due to rejection of religion of their parents. As such their proportions coming down is highly improbable. If you see religious affiliation by age-groups across countries you will find that the 18-24 group has the highest non-religious affiliation and with present statistics available (as old population dies and new replaces) in 2050 there could be 25% non-religious people. However, this still might be an underestimate and as education, knowledge, and prosperousness increases by 2050 the non-religious group may be 50% of world population. I suggest you review your analysis.

    1. Michael Lipka2 years ago


      Thanks for your feedback. These projections do include rates of religious switching in countries where there are reliable data on the topic. This is part of why rising unaffiliated populations are projected in the United States and Europe. Globally, however, these trends are overshadowed by demographic factors in other parts of the world. For more on this topic, see: pewresearch.org/fact-tank/2015/0…

      Thanks again,
      Michael Lipka

  6. Peter James Lineham2 years ago

    We need to have significant cautions about these predictions, for if the New Zealand data is anything to go by, it is based on a very loose quality of empirical research, and the figures are not drawn accurately from the 2006 census as they claim. New Zealand had 57% of Christians in 2001, but in the latest census that figure is 48.5% and this does not inspire confidence in their prediction of 44.7%. They say unaffiliated were 36.6% in 2006 and will be 45.1% in 2050. In NZ unaffiliated needs to include no religion (38.6% in 2013, 32.2% in2006), object to state (4.1% in 2013, 6% in 2006) and should probably include those not filling in the form (8.2% in 2013, 7% in 2006). The figures for other religions are higher than accurate for Islam, low for Hindu. The figures must have been worked off some base that excluded those who objected. This is a disturbing use of statistics flying in the face of all existing research. Peter Lineham, Professor of History Massey University.

    1. Michael Lipka2 years ago


      Thanks for your feedback. To clarify a couple of issues you raise:

      The analysis does not presume to know the religious identity of people who don’t answer religion questions on censuses (including in New Zealand). As a result, estimates are based only on those who respond on the census, recalculated to apply to the entire population. (More detailed explanation here under the subhead “Making Adjustments for Missing Religion Data”: pewforum.org/2015/04/02/appendix….)

      Also, our 2010 estimates differ from the 2006 census figures because we project forward the 2006 data to 2010 before beginning the 2010-2050 projections.

      I hope this helps explain our methodology. Thanks again for reading and for your comment.
      Michael Lipka

  7. vera taylor2 years ago

    I’m guessing that this report will create terror in a lot of Evangelical people. And many will not distinguish between Muslims and radical Islam.

    I think rather than the focus on religion, there should be a focus on the fact that our population will be outgrowing our planet – a fact that is already happening.

    1. Harry2 years ago

      Outstanding point! Too many people are a plague on this planet. I often wonder if this planet is populated with intelligent beings or infested!!

      1. Being human2 years ago

        Don’t worry, you are there, an intelligent being…..

      2. jerry2 years ago

        Hello Harry,

        Would you mind clarifying something for me. You said there are too many people which are a plague on the planet. Are you referring to the the human population as whole are you speaking of individual levels of competence? Any event, the issues we have are small when related to people’s level of intelligence. This planets problem is human population growth. We continue to grow and outgrow what this planet can handle. Intellectual ability is only a means of resolving our current issues. Don’t make the issues about who’s smart and who’s not. Get involved stay involved and together we’ll work it both the smart and smarter.