The age distribution of the church in developed countries is becoming skewed towards the elderly because it is economically rational for potential believers to delay participation in religious activities till they are older. Such is the the claim of a paper by Papyrakis and Selvaretnam, who are economists at the University of East Anglia and St Andrews respectively.

However, empirical evidence, drawn from the same data sources Papyrakis and Selvarentnam used shows that people do not become more religious as they age Marion Burkimsher from the University of Geneva wrote recently "the pattern we see today in western Europe of older people being more religious than younger people is effectively a relic of older generations being more religious in the past and then maintaining their level of religious participation over time." This report is available here

The authors claim that their cost benefit model explains not only the changing age distribution of religiosity, but also the historic reduction in overall levels of observance in the developed world. Their analysis uses a three period model (young, old and afterlife), and allows for members of a population to become religious during the young period, or the old period, and to change their decision at the boundary between the young and the old period. The model uses a number of variables, including the probability that heaven or hell exists, the social and spiritual benefits of religious observance, and the time, and hence income lost, that results from participating in religious activities.

The model under discussion (Click to enlarge

The paper claims implicitly that its results hold true for all religions although it uses explicitly Christian language throughout "Regular church attendance remains very high in countries such as Nigeria, Pakistan, and El Salvador...", and the statistics that are referred to in the section of the paper where the authors present limited empirical evidence for their claims all relate to Christianity. However evidence is emerging which suggests that the decline in attendance which they report has, at least in the UK ceased, and in fact is rising in some denominations (as they acknowledge is true for the Pentecostal community) and is beginning to rise in other denominations.

Church Attendance 2001-2009 (click to enlarge)

Dr Papyrakis, of the School of International Development at UEA said: “Many religions and societies link to some degree the cumulative amount of religious effort to benefits in the afterlife. We show that higher life expectancy discounts expected benefits in the afterlife and is therefore likely to lead to postponement of religiosity, without necessarily jeopardising benefits in the afterlife." However because their model considers the decisions of hypothetical individuals, and the effect of the increased life expectancy which is found in developed countries relative to less developed ones, the paper cannot without further work define an end state for the religious community, or even the age distribution which one would expect to find. Were the findings of this paper true, it would be expected that in developed countries the age at which people first become religiously observant would be likely to show a marked skew towards the older generations. However several studies show that this is not the case as a report from the Barna Group, a leading US research organisation shows: "The current Barna study indicates that nearly half of all Americans who accept Jesus Christ as their savior do so before reaching the age of 13 (43%), and that two out of three born again Christians (64%) made that commitment to Christ before their 18th birthday. One out of eight born again people (13%) made their profession of faith while 18 to 21 years old. Less than one out of every four born again Christians (23%) embraced Christ after their twenty-first birthday. Barna noted that these figures are consistent with similar studies it has conducted during the past twenty years."

It would also be expected that where there was a significant disutility to practicing a religion for reasons of persecution, for example the early church or early non-conformist communities in England that religious participation (either among the young or the old), would be suppressed, and this paper makes this point. In illustration of this argument they cite a paper: "e.g. in the early Christian Church), these costs are particularly high (see Bruce, 1993 for a discussion) and may hence discourage religious participation." However examination of Bruce's work shows that he argues that the economic model for restriction of church growth is flawed:

"The early Christian Church had very high start-up costs: Christians were persecuted, and some were used as lion feed. The early English Quakers were persecuted but thrived ... At this point in the argument it is enough to note that there appears to be no universal connection between 'costs' and adherence."

Dr Papyrakis' said in conclusion: “To increase overall attendance, religious establishments should aim to reduce any discomfort of entry to religious newcomers, both old and young. This may involve making information about the organisation easily accessible to them and helping new-comers to follow religious activities without feeling lost or uncomfortable." He went on to suggest that emphasising the socio-economic and spiritual benefits that religious observance brings would counterbalance the negative impact of life expectancy on religiosity. As conclusions these are substantially anodyne. It makes intuitive sense for any Christian church to do all he suggests, but his conclusions are so general that it is difficult to attach them to the findings of his model. His conclusions also fail to take account of the fact that the demographics of many developed countries are such that there is an increasing proportion of elderly people in the population, and as such even if churches are perfectly representative of the population there would be a skewed age distribution within them.