Page 98 - Invited Paper Session (IPS) - Volume 1

IPS102 Arjan B.
sheets in the SNA. The IIWS is used for Currency and Deposits, and Loans. Data
gaps between micro and macro totals are allocated as much as possible to the
right individuals, households, or groups. For instance, the coverage of loans in
the IIWS is incomplete. Assets related to mortgages with a savings or
investment component are not observed, while they are included in the
National Accounts. We allocate part of the data gap to mortgage-payers.
Another gap occurs because the loans of the self-employed are netted in their
business value, while these are made explicit in the SNA. This part of the data
gap is estimated by the interest receipts and payments of self-employed and
allocated accordingly. We use the macro data from custodians to estimate
debt securities and equity and investment fund shares, but use the IIWS to
break this down over households. The other accounts receivable/payable are
allocated over individuals in relation to their current accounts.
The IIWS does not included data on pension entitlements. We use the PCS,
which aims to give an overview of the mandatory (employment related)
pension entitlements of the population. We use the accrued to date
entitlements to calculate the net present value (NPV) of the future benefits
2
receivable. For each individual we combine the PCS with their life expectancy
and legal retirement age. Using the Interest Rate Term Structure (IRTS) for the
3
discount rate, as prescribed by the supervisor , we calculate the NPV of these
entitlements. This calculation does not lead to a perfect match with the SNA
totals for some reasons. First, we used the PCS of 2014 for which we assumed
developments to 2015 and 2016. Second, the PCS is a sample survey, taken
from pension funds, and we need to impute the missing individuals. Third,
these entitlements are old age only and partner pensions and disability
pensions are excluded. Finally, pension funds could have used other
parameters for life expectancy. Using for example socio-economic class could
possibly lead to different outcomes. The remaining difference between our
estimate and the NA total is allocated proportionally over the individuals. This
approach differs from Saez and Zucman (2016), who impute the pension
entitlements for 60% to retirees and the remaining 40% to workers. With our
method we find that these shares are approximately the other way around.
We further include the }IFCS of 2014 for the voluntary pension
entitlements. The }IFCS’ main aim is to gather micro-level structural
information on euro area households' assets and liabilities. The survey also
collects other information in order to analyse the economic decisions taken by
households. Only private households and their current members residing in
the national territory at the time of data collection are included in the survey,
which is held once every three years. The HFCS gives us an average amount of

2 Life expectancy is imputed in the data set for each combination of age and gender
3 The Dutch Central Bank prescribes the IRTS monthly, see: https://statistiek.dnb.nl/
87 | I S I W S C 2 0 1 9