administrative periodic inspection ii

RiskCo has designed a standard approach and calculation model for the Administrative Periodic Inspection. This enables us to quickly make all of the necessary calculations for all members of your database for their entire period of participation. The API is an independent check of the quality of your pension administration. You can immediately see all of the incorrect and missing data in your administration. Apart from data cleansing, the calculation model designed during the API can also be used for future audits, scenario calculations and calculations of liabilities. Because the API is a standardized process, you will save on external costs.

How does an API work?

  • We receive a copy of your administration database and interpret the raw data. We check for completeness and consistency and we may even already find the first problems in your data.
  • We adjust our blueprint model according to your regulations, board decisions, etc. and connect our model to the structure of your administration database.
  • We calculate all of the policies.
  • The results are compared to the expected results provided by you (for instance taken from the annual report or communication to participants).
  • We analyze all differences. We first check whether any mistakes were made on our part. If we made no mistakes, we then search for the cause of the inconsistency.
  • All participants with the same cause are grouped together. The total inconsistency for all members in this group is an indication of the severity of the problem.
  • We continue until the calculations for all members are exactly right, or when all of the members can be put into at least one group.

In conclusion, we make a report of all of the problems and their severity and we also suggest how these problems can be solved.

administrative periodic inspection ii

During an API, more problems are found than during a sample test

Many pension funds or insurance companies use a sample of their database to examine their administration. But a sample never offers the certainty that an API will give you.

The chance that you will find all of the problems by using a sample is virtually zero. What will you find during an API that you will probably not find during a sample test?

  • Wrong calculations for groups of members that are not represented in the sample.
  • Calculations that are internally known but are not made according to the official regulations.
  • Data errors for individuals.
  • Members that are not supposed to be in the administration but are, and vice versa.
  • Ambiguity in the regulation documentation that may lead to errors.
  • Shortcomings in the documentation on the administration.

After the API you have a clear and complete picture of the quality of your administration. You know which issues are present and what the financial consequences are. When we are done, there will be no more surprises.

during-an-api-more-problems
sample-test

Generally used sample tests

A common sampling technique is as follows: For each pension fund, the entitlements of 30 participants are manually recalculated from commencement to conclusion and compared to the value recorded in the administration system. 15 participants are drawn randomly from all participants in the administration. The other 15 are obtained by selecting 3 participants from the following groups: divorce, exchange of pension (old age pension) , death in active employment, disability and value transfer of pension rights.