BSPS Prize 2017
The winner of the 2017 BSPS Prize for the best submitted Masters dissertation on a demographic topic from 2016 was announced at the 2017 BSPS Conference. The Prize was awarded to Ingrid Amalia Ceballos Gaystardo (University of Southampton), for her dissertation Revisiting the Hispanic paradox - age-gender variation patterns in hypertension and diabetes across the foreign-born and U.S.-born Mexicans.
From the judges' comments:
This is an excellent dissertation, well-written, accessible and displaying very competent analytical skills. The introduction was a particularly strong feature, providing a very helpful contextualisation with respect to both the US healthcare system, and models of migrant health.
Congratulations to Ingrid Amalia.
BSPS response to ONS consultation on changes to ONS products
BSPS thanks all those members who responded to the Vice-President's call for input. The BSPS response was sent to ONS on 8 February 2016 and the full response can be accessed here:
BSPS Response to ONS Consultation on changes to ONS products
BSPS submission to PASC on migration statistics
The BSPS submission to the Public Administration Select Committee (PASC) Programme of Work on Statistics has now been published on the Parliamentary website, where it can be viewed with other submissions:
The BSPS submission can also be downloaded as a PDF here:
BSPS submission to PASC- Migration statistics
as agreed by BSPS Council at their Council meeting on 13th January 2012
At its meeting on 14th October 2011 Council noted the absence of a documented set of procedures for membership matters. Clause 6 of the Society’s Constitution gives responsibility for decisions on membership to Council. It is therefore a matter of ensuring that the necessary procedures for Council’s handling of membership matters are documented in sufficient detail. It was agreed that Council would consider details of a membership procedures document at its next meeting with specific reference to,
1. Content of membership application form and information required by Council to determine membership applications;
2. Procedures for reviewing and taking action against any member of the Society;
3. Categories of membership.
1. Content of membership application form and information required by Council to determine membership applications
Information requested on the current membership application form comprises,
• Name and contact details;
• Occupation and organisation affiliation;
• Areas of demographic interest – by ticking any number of 35 categories plus an ‘Other’ category for writing in’s.
Information presented to Council for decisions on membership applications comprises name, occupation and organisation affiliation.
By far the most significant affiliations of members of BSPS are public service organisations including universities, local authorities, ONS and other Government departments and agencies. Most remaining members are either academics at overseas institutions or self-employed UK researchers and consultants, with members in each of these categories having an ongoing working association with other BSPS members. There are a limited number of current members who do not fall into the above categories. Most applicants for membership of the Society are also known to a member of Council, who is able to endorse the credentials of the applicant in terms of ‘a scientific interest in the study of human populations’ (Clause 1, BSPS Constitution).
Against this background it would be inappropriate to undertake a general review and expansion of the information required on the membership application form. Rather, where Council considers that it cannot verify the scientific demographic credentials of an applicant, it should defer consideration of the application pending receipt of further information on the applicant’s demographic interest and background and their reasons for applying.
To ensure that this process is transparent the membership application form should be amended to additionally read,
‘At its discretion, BSPS Council may request further information from any applicant to assist its consideration of any application for membership. Membership applications, and any requested additional information, will be considered by BSPS Council at its next available meeting.’ This has now been added.
2. Procedures for reviewing and taking action against any member of the Society
Currently BSPS does not have any procedure for taking action against any member of the Society. This is due to the fact that such an eventuality has not previously arisen throughout the history of the Society. However, it would be appropriate to consider the matter at this time alongside consideration of the membership application procedures. Key considerations for reviewing or taking action against a member would comprise,
• Trigger for Action: a written complaint from a BSPS member(s) which included at least one of the following grounds for action,
Non-payment of subscription for previous year
Mis-use of BSPS materials
Mis-representation of participation in BSPS events and affairs
Mis-representation of BSPS membership as an expert qualification
Actions not in accord with BSPS aims and objectives
• Consideration of Action: by BSPS Council at its next available meeting. The affected member to submit a written account if desired.
• Sanctions to be Applied: a range of sanctions to reflect the findings of Council to comprise,
An informal conversation with a member of Council;
A written warning signed by at least 2 officers of BSPS;
Revocation of BSPS membership without refund. If membership has been revoked, re-application for membership may be made in the following calendar year.
3. Categories of membership
Most members of BSPS subscribe on an individual basis. However, BSPS does offer a corporate membership category. Currently 10 organisations, dominantly in the local authority sector, subscribe through this category. Corporate membership confers little advantage to either the organisations or the individuals concerned.
Adoption of the above suggested membership procedures could make it difficult to determine applications for corporate membership or to take action against an individual affiliated to a corporate member. It is therefore recommended that the Society no longer offers a corporate membership category. This may cause some difficulty to a limited number of existing BSPS members. But, it would enable the proposed procedures to be applied more easily if all were individual members of the Society.
The above membership procedures were agreed and collated into this document. New members will be advised of the document upon confirmation of membership.
Call for Evidence by the House of Lords Committee on Public Service and Demographic Change
Submission from the British Society for Population Studies, August 31st 2012
1. The British Society for Population Studies (BSPS) comprises persons with a scientific interest in the study of human populations. Its main objectives are to further the scientific study of biological, economic, historical, medical, social and other disciplines connected with human populations and to contribute to the public awareness of them. Our four hundred BSPS members include the public sector (including central and local government and health services), academia and business. The Society discusses current scientific research at its annual conference which this year meets on September 10-12. Among its 120 presentations are thirteen contributions on ageing which investigate a broad range of impacts on health, social care, families, households and economics.
2. The ageing of the UK population has been achieved through a long-term reduction of mortality that first allowed more people to survive to adulthood and then increased the absolute number of elderly people, along with a reduction of fertility that has shifted the age distribution away from younger people.
3. Ageing has had tremendous impact on UK society in the past fifty years. The strong expectation of living to 60, and the likelihood of living to 80, has given confidence to young adults and given rise to longer independent lives after children have left home. With the separation of generations, whose integration in society has taken place within different contexts and events, age now marks great differences in society. Since society’s traditions and expectations change more rapidly and within a longer lifetime, the experience of the elderly is often seen as less relevant and therefore devalued. The near guarantee of surviving parenthood has contributed to the growth of the nuclear family, and grandparents live in greater isolation, on average.
4. The current discussion of the costs of ageing should be taken in the context of these changes, and within social aims which could encourage integration and cohesion across all ages.
5. Although the committee has focussed on ageing for its discussion of the impact on public services of demographic change, it should be aware that the population’s distribution within the UK and the number and structure of households are also changing, with very significant impact on public services.
6. This submission from the British Society for Population Studies does not attempt to cover all aspects of ageing and public services, nor all the questions posed by the Committee. It addresses some of the issues of measuring population change that the committee has raised and that our members have been able to respond to in the short timescale allowed by the Committee: measurement of ageing; the certainty with which projections of ageing are made; the limits to longevity and the influence of obesity within this; and the causes of the predicted growth in the elderly population, which is particularly important to understand when planning for different time-scales of the future. Inevitably we have identified gaps in knowledge where investment would prove worthwhile. We would be happy to respond to further questions that the Committee may have.
Measurement of ageing
7. Both ageing as a shift in population distribution towards older age groups, and the absolute numbers of older persons, do matter.
8. Ageing may in particular affect macro economic considerations The predominant view is that ageing will support lower economic growth because it “squeezes” the 20-64 age group, leading to low (or negative) rates of labour force growth. A high-skilled growing labour force is generally needed for sustained economic growth, which may be a government aim. In addition, population ageing causes a shift away from investment to consumption since government expenditure tends to increase to meet the increased demand for pensions and other old age-related benefits (such as health care).
9. The absolute numbers of older people matters because local agencies (health or local authority) need to deliver services and care to frail older people.
10. When measuring ageing – the shift in age distribution – care must be taken to distinguish measures such as the Old Age Dependency Ratio (ODR) and the Old Age Support Ratio (OSR, the reciprocal of the ODR). Expectations of change should use the same numerator and denominator, in particular referring to the same age groups, in order to be comparable. For example, Lord Bichard said “We have talked a lot about increasing numbers of older people, but the old age support ratio diminishes only slightly. It goes from 3.2 now (2010) to 2.9 in 2035” (House of Lords 2012, Transcript Q50), quoting ONS (2011a, p.6). In this case, the OSR was measured as ratios of the population aged 16 to pensionable age to the population of pensionable age and over. Under the Pensions Act of 2011 pensionable age thresholds will rise decade by decade and there is a note that the timetable may be speeded up. So the smallness of the fall on which Lord Bichard remarked is a direct consequence of legislation and implementation of policy. If a fixed age threshold had been used, such as age 65, the OSR for the UK would have been 3.9 in 2010 and 2.6 in 2035 (based on the ONS 2010 principal projection).
The certainty with which projections of ageing are made
11. Measures of certainty for population projections are not yet an exact science. When attempted, they distinguish the uncertainty attached to each of the components of future population change – primarily fertility, mortality and migration. A quantified level of uncertainty can be based on expert opinion of future ranges of population change, or on the past instability of trends (measured using statistical time series models), or on the record of past forecasts.
12. In the UK, Chris Shaw of ONS led a project using expert opinion to set the uncertainty around each component of future UK population change, from the 2006-based national population projections (Shaw 2007). The project has not reached completion with publication, but from a progress report (Rowan and Wright 2010), we can see that the confidence bands are widest for the youngest ages and diminish towards the older ages. Phil Rees has computed the ratio between the 95% confidence intervals and the projected populations, finding that the percent uncertainty diminishes from 13% for ages 0-4 in 2031 to 1% for ages 50-69, grows to 3% for ages 75-79 but then increases considerably at very old ages to be 64% for the relatively small population aged 100+. In other words it is 95% likely that the 2031 population of centenarians would be up to but not more than 32% either side of the official 2006-based projection. Errors we make in forecasting mortality rates pile up in the oldest old ages.
13. However, when we compare these 2006-based projections with the most recent projection based on mid-2010 populations, we see that the 2010 based projected population falls above the upper confidence level for each of children, working ages and elderly For the population aged 65+, the 95% confidence interval is 2% either side of 2006-based projection, but the 2010 projection is 3% above it, outside the confidence interval.
14. We cannot therefore be very confident in the confidence intervals calculated so far. The calculations use errors for fertility, mortality and migration averaged from those provided by members of the National Population Projections Expert Panel. But the opinions offered were few and therefore the estimate of the errors very uncertain. What this discussion indicates is that ONS needs to invest further in methods for estimating confidence levels in their projections, if they are to answer the reasonable questions made of them.
Trends in longevity
15. There are two polar views on this issue. The first, put forward by Jay Olshansky and colleagues (e.g. Olshansky 2001) is that we are approaching the limits to life expectancy and that a number of disease trends (e.g. increasing obesity leading to much higher rates of diabetes and associated mortality) will mean that we will not see the continuation in improvement in mortality rates at older ages. The second, put forward by James Vaupel and colleagues (e.g. Oeppen and Vaupel 2002) is that the historical record of the countries with the best life expectancy records suggests no limits to improvements driven by progress in well-being and medical science.
16. A summary of current expert thinking on the potential impact of obesity on projections of longevity can be found in the discussion of the National Population Projections Expert Advisory Group which helped inform the mortality assumptions used in the 2010-based UK population projections published by the ONS (2011b). The panel generally agreed that in the UK obesity is likely to have a greater impact on morbidity than on mortality, but there was no clear consensus on its impact. Much will depend on future changes in behaviour which could increase or reduce the incidence of obesity and the age at which people become obese, and whether obesity is regularly medically monitored, as seems to be the case in the USA.
17. The report of the Foresight programme on Tackling Obesities (Government Office for Science 2007a,b) estimated that if obesity levels increase from 2010 as they predict then by 2050 female life expectancy would be lower by a fifth of a year while male life expectancy would be lower by a third of a year compared to what it would have been without this rise. This though has to be set in the context that female and male period life expectancies at birth are projected to increase by over 6 years and 7 years respectively. Thus obesity will have perhaps a surprisingly small impact on life expectancy at birth, due mainly to greater reductions in mortality from other, non-obesity related causes. The impact on longevity over a shorter, say 10 – 15 year, time span will be much smaller. It should be stressed that these trends depend on current policy of investing in better survival and on continuing failure to stop the increase in obesity.
18. Just as obesity could potentially have an impact on longevity and health care costs so could other emerging health threats such as pandemic flu or the resurgence of TB. While the likelihood of these occurring may be small the impact could be significant. The impact of major epidemics can be allowed for in population projections but the timing of such events is unknowable.
What causes the predicted growth in the elderly population?
19. The increase in the number of old people is not just a function of improving mortality before and during old age, but it also depends on the size of new cohorts entering old age (e.g. having their 65th birthday) compared with those who went before them. To what extent does the future population of the old grow because (a) mortality is declining, because (b) the birth cohorts entering old age in the future are larger than they have been in the past, or because of (c) the impact of international migration of older people? This is quite a difficult question to answer because it involves designing the correct scenario projections with a careful age-period-cohort analysis, which has not been done. However, the question is an important one to answer. An initial estimate has been made using the scenarios for UK population projections developed at Leeds University (Rees et al., 2012a,b).
20. The effect on the population of projected reduced mortality, adds 3.3 million to the UK population between 2001 and 2026. Making the approximation that all these extra persons are old people (aged 65+), one can subtract this mortality effect from the total population change of 6.1 million to give an approximate estimate of changing cohort size of 2.8 million in 2001-2026.
21. Thus the percentage contribution to a growing elderly population of improving longevity is 54% and of cohort replacement is 46% during this period. The estimates also assume that the net impact of migration is negligible compared to these two other effects: ONS expects a net annual emigration of 2-3 thousand, about 0.1 million over twenty five years. This, and the impact of net immigration at younger ages, will be part of the estimated effect of changing cohort size.
22. Thus it is likely that the impact of longevity accounts for only about half of the growth in the elderly population over the period 2001-26. The effect of changing cohort size, reflecting mainly the baby boom in the UK in the 1950s and early 1960s, contributes the rest of the growth. However, as the smaller birth cohorts born since the mid-1960s reach older age, their impact will be to reduce the size of the elderly population, partly countering the impact of longevity. This is the main reason that ONS projects less growth for the total elderly population aged 65+ 2040-2060 (161 thousand per annum) than 2010-2030 (261 thousand per annum).
23. The significant immigration of the 1960s will have reinforced and lengthened the baby boom effect on the current elderly, while the rise in immigration from mid 1990s will help to offset the lower fertility of the 1970s. A more precise disaggregation of the reasons for change in the elderly population in each of the next few decades would be of use to the Committee’s deliberations.
24. Both the changing absolute size of the elderly population and its size relative to other age groups will have a great impact on the nature of public services in the next decades. The dynamics of this change are understood but could usefully be measured by government with greater clarity at both national and subnational scales.
Government Office for Science (2007a). FORESIGHT. Tackling Obesities: Future Choices –Project Report. 2nd Edition. Department of Innovation, Universities and Skills, http://www.foresight.gov.uk/Obesity/Obesity_final_part1.pdf.
Government Office for Science (2007b). FORESIGHT. Tackling Obesities: Future Choices –Modelling Future Trends in Obesity & Their Impact on Health. 2nd Edition. Department of Innovation, Universities and Skills, http://www.bis.gov.uk/assets/foresight/docs/obesity/14.pdf.
House of Lords (2012) Call for evidence on the impact of demographic change on public services, Committee on Public Service and Demographic Change. Oral hearings, 5 July 2012,Transcript, http://www.parliament.uk/business/committees/committees-a-z/lords-select/public-services-committee/, Session video, http://www.parliamentlive.tv/Main/Player.aspx?meetingId=11167.
Oeppen J, Vaupel JW. (2002) Broken limits to life expectancy. Science, 296: 1029–1031.
Olshansky SJ, Carnes BA, Désesguelles A. (2001) Prospects for longevity. Science, 291: 1491–1492.
ONS (2011a) National Population Projections, 2010-based. Office for National Statistics, Statistical Bulletin, http://www.ons.gov.uk/ons/rel/npp/national-population-projections/2010-based-projections/index.html
ONS (2011b) A note of the meeting of the national population projections expert advisory group. http://www.ons.gov.uk/ons/rel/npp/national-population-projections/2010-based-projections/rep-2010-based-npp.html#tab-Appendix-A--Note-of-the-meeting-of-the-national-population-projections-expert-advisory-group
Rees, P., Wohland, P., Norman, P. and Boden, P. (2012a) Ethnic population how accurate have they been? Population Trends 128: 8-23.
Rowan, S. and Wright, Emma (2010) Developing stochastic population forecasts for the United Kingdom: Progress report and plans for future work, Office for National Statistics, United Kingdom. Paper presented at Session 13, Item 13 – Stochastic national demographic projections, Conference of European Statisticians, Joint Eurostat/UNECE Work Session on Demographic Projections, 28-30 April 2010, Lisbon, Portugal, http://www.unece.org/stats/documents/ece/ces/ge.11/2010/wp.40.e.pdf.
January 2011: Reductions in government expenditure: BSPS statement following a members' survey.
The government's choice to severely reduce public expenditure will affect the availability and analysis of population data. BSPS is concerned at the impact of these reductions on its members, and more widely on the scientific study of human populations.
A BSPS survey of members in December 2010 helped BSPS to respond to ONS' consultation on its work programme. The response highlighted the primacy of the Census, and the welcome and important growth recently in robust government statistics for small areas. It also highlighted ONS outputs that 'are not directly aimed at specific policies but are essential for academic and other fundamental investigations. Members valued the availability of survey microdata and detailed sub-national statistics for these reasons'.
The BSPS survey also reviewed reductions in demographic statistics outside ONS, expressing concern about the impact of reduced budgets in the coming financial year.
Report to Council and decisions.
BSPS asks members and others to keep it informed of the impact of changed government expenditure on the availability & analysis of demographic statistics and more generally on population studies.