Page 298 - Invited Paper Session (IPS) - Volume 1
P. 298

IPS153 John D.
                  1.  Introduction
                      For  countries  that  do  not  have  a  Central  Population  Register  (CPR)  on
                  which demographic statistics can be compiled, the production of coherent
                  statistics  on  population  stocks  and  flows  can  prove  challenging.  This  is
                  particularly  true  for  those  countries  that  have  relatively  high  and  variable
                  migration flows that are difficult to estimate.
                      The traditional approach to population estimates in these scenarios is an
                   application of the demographic component or cohort component method to
                   postcensal  population  estimates  and  then  a  recalibration  of  population
                   estimates  for  intercensal  estimates.  In  a  Eurostat  review  of  31  countries
                   (EUROSTAT, 2003) published in 2003, 19 countries were identified as using the
                   component method for population estimates. This approach can be labelled a
                   flows first based approach in that the population flows are first estimated and
                   then used to derive the stocks based estimate. The weakness with this approach
                   is that any errors or bias in estimating the components of population change
                   (births, deaths, immigration, emigration) will be carried forward from timepoint
                   to timepoint. These concerns, amplified in the presence of high migration flows,
                   are one of the reasons why some countries, such as Ireland, undertake a census
                   at  5  yearly  intervals.  The  census  provides  a  benchmark  to  recalibrate  the
                   population  estimates  at  regular  intervals.  In  a  subsequent  review  in  2015
                   (EUROSTAT, 2015), Eurostat found that 31 of 44 countries depended on the
                   census  for  annual  population  estimates  and  of  these  31  countries  only  9
                   supplemented their population estimates with information from registers.
                      For  many  countries,  the  census  is  the  backbone  to  the  system  of
                   population estimates. Conducting a traditional census is a major logistical
                   exercise presenting many challenges, including those of cost, logistics and
                   timeliness. Given the associated challenges, these countries are seeking to
                   modernise and use administrative data as the basis of a modern system of
                   population statistics.
                      In the 2000 round of censuses only 4 of 44 countries conducted a register-
                   based  census  where  the  enumeration  was  based  on  a  Central  Population
                   Register  (CPR)  (UN-ECE,  2008).  In  a  register  based  census  persons  and
                   households are simply counted using the CPR. The CPR, available in some
                   national administrations, is the backbone to the co-ordination and delivery of
                   public services to individuals and households.
                      The ability to compile reliable population estimates from administrative
                   data  sources  is  a  significant  milestone  in  transitioning  from  a  traditional
                   system of population statistics (with the traditional census at its core) to a
                   modern  system  of  social  statistics  with  administrative  registers  and
                   administrative data at its core. A modern system also holds the promise of
                   being able to conduct a census on an annual basis at a fraction of the cost of
                   what a traditional census would cost. Reliable annual population estimates

                                                                     287 | I S I   W S C   2 0 1 9
   293   294   295   296   297   298   299   300   301   302   303