Irish Population Estimates from Administrative Data Sources, 2020

This chapter showcases the potential of administrative data to produce population statistics like some of those obtained in the census of population. Many of the underlying trends that are seen in the census are also seen in these statistics. In some cases, there are notable differences between what has been produced here and what we see in the census and other surveys. This mainly happens because the method of data collection is very different.

The experimental methodology used to compile the following statistics is outlined in greater detail in the Methodology chapter. Using this methodology, the population of Ireland was estimated to be 5.2 million in April 2020. This estimate is produced using data collected from administrative records only and applying a set of ‘rules’ to decide whether a person is likely to be usually resident or not. Applying a population concept of usual residence to data collected from administrative records is challenging and the CSO are undertaking further analysis to improve the accuracy of these estimates. It may take several years of refinement and commitment across the broader system to adopt the National Data Infrastructure, before the CSO have sufficient confidence in these estimates so that they may play an integral part in the production of official statistics.

Age structure of the population

The number of males and females in 2020, by single year of age, is represented in the population pyramid in Figure 2.1. There were high numbers of births in the late 1970s and early 1980s. The children born then were in their late 30s and early 40s in 2020 and are the main contribution to the bulge in the middle of the population pyramid. Another contribution to this cohort is as a result of strong inward migration of 20 to 30 year olds starting in the late 1990s up until 2009. Even though the number of births peaked at 74,000 in 1980 the number of persons aged around 40 in 2020 is estimated to be about 88,000 reflecting the net inward migration. This cohort then drove the high birth rates from 2007 on, peaking at over 75,000 in 2009.

Low birth rates in the late 1980s and 1990s combined with net outward migration between 2010 and 2014 has resulted in the smaller population of persons in their teenage years and 20s. As a result, we have seen a drop off in births recently reflected in the shape of the pyramid from age 10 and under.

Age structure by county

The interactive population pyramid below shows how the age structure of the population differs across Ireland. In very urban areas such as Dublin City, Galway City and Cork City there are relatively fewer children and large cohorts of young adults. Whereas in more suburban and rural areas there are higher proportions of children and adults in their 30s and 40s. This difference in the population structure is particularly noticeable when comparing Cork City and Cork County for example.

Carlow Dublin City Dún Laoghaire–Rathdown Fingal South Dublin Kildare Kilkenny Laois Longford Louth Meath Offaly Westmeath Wexford Wicklow Clare Cork City Cork County Kerry Limerick City and County Tipperary Waterford City and County Galway City Galway County Leitrim Mayo Roscommon Sligo Cavan Donegal Monaghan

Visit Table IPEADS02 on PxStat

Age dependency

Dependents are defined for statistical purposes as people outside the normal working age of 15-64. Dependency ratios are used to give a useful indication of the age structure of a population with young (0-14) and old (65+) shown as a percentage of the population of working age (15-64).

The total dependency ratio stood at 50.8% for the State. When examined by gender the results show total dependency was higher for women, at 51.6%, than for men at 50.1%, this difference is driven by the gap in the old dependency ratio for women (22.6%) compared with men (20.2%). The young dependency ratio is similar for women and men.

Young dependency, shown in Map 2.1, is the number of young people aged 0-14 as a percentage of the population of working age, stood at 29.4% for the State. Cork and Dublin Cities had the lowest young dependency at 20.2%, followed by Galway City at 23.4%. Laois had the highest young dependency at 35.3%, followed by Meath at 34.4%.

Old dependency, the number of older people aged 65+ as a percentage of the population of working age, stood at 21.4% for the State. Map 2.2 shows that counties of the North West, Kerry and Dún Laoghaire-Rathdown recorded the highest old dependency ratios between 26.2% and 29.0%. Fingal (14.8) and Galway City (16.4) recorded the lowest old dependency ratios.

Average age

The average age of the population is estimated to be 37.9 years in 2020. Fingal recorded the youngest population, followed by Kildare. Mayo and Kerry have the oldest populations, followed closely by Dún Laoghaire-Rathdown. Table 2.1 shows the top and bottom average age of the population for the county and city administrative areas by sex.

At 38.4, the average age for women was one year older than men who had an average age of 37.4. When examined by sex, the average age for women is higher in all county and city administrative areas.

Marital status

Martial status data can be derived from information recorded by the Department of Social Protection (DSP). Figure 2.3 below shows marital status by age and sex for all persons aged 15 years and over. The chart shows how the number of single persons decreases with age. Over 60% of women between the ages of 35 and 44 are married compared with 53% of men. The proportion of separated and divorced persons is highest between the ages of 55 and 64 for women at 11% while for men it peaks at 8% between the ages of 65 and 74.

Nationality

Data on nationality can be derived from information collected by the DSP and provides good coverage across the population. It is important to note that for many individuals this data may have been collected several years ago and in some cases, people may no longer identify with the nationality recorded here. In particular, many people have become Irish citizens by naturalisation in the last 10 years or so. This more recent status may not be reflected in these statistics.

The age and sex breakdown of a selection of nationalities can be seen in Figure 2.4 below. The buttons below the chart can be used to view the age structure of each nationality. The age profile of different nationalities varies widely. For example, a large proportion (39.0%) of Australian nationals are under the age of 15. Brazilian nationals are concentrated in the 25 to 34 age group (56.6%) while at 44.5%, Polish nationals are more likely to be aged between 35 and 44 and UK nationals are mostly over 45 years of age (56.8%).

Australian Bulgarian Brazilian Chinese Czech German Spanish French UK Croatian Hungarian Irish Indian Italian Lithuanian Latvian Nigerian Dutch Other EU28 Filipino Pakistani Polish Portuguese Romanian Slovak American (US) South African

Visit Table IPEADS04 on PxStat

Families

Family data can be derived from relationship data recorded by the DSP. This requires transforming data that identifies if persons are in cohabiting couples, married couples and/or parent/child relationships in to family types. It is then necessary to determine if these families live in the same household. Figure 2.5 shows the age profile of children in these different family types. Cohabiting couples with children are most likely to have all children under age 15 at 74.8% compared to married couples with children (50.6%). The age profile of the children in one parent families differs depending on whether the parent is the children’s mother or father. Fathers are most likely to be living with older children; one parent father families record 63.3% of all children aged 15 years and over compared to one parent mother families at 31.5%.

Principal Economic Status

Deriving data on the economic status of persons aged 15 and over can be attempted by looking at what datasets they appear on and/or by their type of tax return or welfare payment. It can be difficult to confidently ascertain what the likely ‘principal’ economic status is however, particularly in cases where someone is appearing on multiple different datasets. For example, some students may also be working, and it may not be clear whether they are working part time or studying part time. This methodology is quite different to the official CSO estimates on economic status reported in Census and the Labour Force Survey publications, which are based on a person's self-assessment of their economic status, see also Background Notes.

The level of unemployment will reflect the situation pre-global COVID-19 pandemic as persons in employment in the months prior to receiving a PUP payment will be included as ‘persons at work’.

Figure 2.6 gives a breakdown of principal economic status by sex for persons aged 15 and over. For this publication, the labour force is comprised of persons at work and all unemployed persons. The results show that the total in the labour force in April 2020 stood at 2,738,323, which represents 65.4% of all persons aged 15 years and over.

(Note: The official labour force and unemployment estimates are compiled in the Labour Force Survey (LFS). The results in this report differ for methodological reasons from these official estimates. See Background Notes.)

In Figure 2.7 we can see the proportion of persons at work and unemployed in each county.

NACE Sector

Data on industry group is provided from the PAYE Modernisation (PMOD) and Form 11 Income Tax returns (ITForm11) data sources and is coded using NACE – the Statistical Classification of Economic Activities in the European Community. In the Census the industrial group of each person is determined from a question requesting details of the business of the person's main employer. Using only administrative data sources to determine each person's industry group presents a number of possible issues;

• a person may have a number of employers/industry groups with different NACE codes on PMOD and ITForm11
• a person may change employers/industry during the reference period
• where an employer provides ancillary services, the person working in that service should be recorded to the NACE code associated to that service, however PMOD and ITForm11 may classify such persons to the main industry NACE code of the employer

Figure 2.8 shows the numbers at work by industry group and by sex highlighting the differences between male and female employment. A significantly higher number of men worked in construction compared with women. By comparison more women than men worked in health and social work. Public administration and defence also shows higher numbers of women than men, this group includes administration in state services including health, education and social services as well as defence and fire services.

Electoral Divisions

A key objective of a population count from administrative data sources is the collection of data at more granular levels of geography. Up to date and accurately geocoded address data is needed to produce statistics for small areas, electoral divisions and other geographical boundaries below county level.

Eircode data is an important part of this process and Eircode coverage on administrative data is improving all the time but the pace of adoption is perhaps an area for development. In many existing datasets however Eircode coverage is quite poor. The CSO have been examining different ways to improve the quality and coverage of geocoded address data. The best way forward is for public sector bodies to collect and capture the Eircode on their data holdings. Below is a sample of the type of statistics that could be produced for electoral divisions and other sub county geographies. Further work is needed to get to a position where we can publish statistics for the full set of electoral divisions. See Methodology and Background Notes.

Map 2.3 shows the average age of the population for each electoral division. Bunaveela, Co. Mayo (53.6) and Coos, Co. Galway (53.3) record the highest average ages, while Kilbarry, Co. Waterford (28.0) and The Ward, Co. Dublin (29.0) had the lowest average ages.

Figure 2.9 shows population pyramids for a selection of electoral divisions from different provinces of the country illustrating how age structure of the population differs across electoral divisions.

Blarney Donegal Bearna Kilkenny No.1 Urban

Table 2.2 shows a breakdown of selected electoral divisions by marital status, economic status and nationality and includes details of these attributes for the State.