1. Introduction
The Dutch pension landscape is undergoing a profound shift from Defined Benefit (DB) to Defined Contribution (DC) under the new pension rules. Financial, actuarial and economic risks are reflected more directly in individual pension outcomes.
Participants of the same age can nevertheless have materially different financial positions due to differences in entry age, contribution history, salary development and accumulated capital. A collective pension ambition therefore does not necessarily indicate whether an individual participant is on track to achieve a relevant pension objective.
A goal-oriented approach addresses this by defining a participant-specific pension target in real terms or as a target annuity. This target does not replace the collective pension ambition. It translates that ambition into a participant-oriented framework.
This paper introduces Targeted Defined Contribution™ (TDC) as a goal-oriented layer within the existing pension and ALM framework. TDC links the pension objective to the participant's financial position, required return, probability of success and risk tolerance, and uses these variables to assess the appropriate investment strategy within the applicable collective investment constraints.
TDC is based on seven principles:
- Goal-based, liability-aware risk allocation based on a pension objective or notional pension liability;
- Participant-centric analysis incorporating financial, actuarial, salary and career-related information;
- Distance-to-target measurement to identify differences between same-aged participants;
- Dynamic adjustment as the participant's financial position changes;
- Integrated accumulation framework throughout the investment horizon;
- Risk-tolerance alignment linking the pension target, probability of success and the risk tolerance;
- Objective-first decision-making, using the pension objective as the primary economic reference point.
The economic optimisation problem can be stated as follows: given a defined pension target, the participant’s financial position, the remaining investment horizon and the applicable collective constraints, identify an investment strategy that provides an appropriate probability of achieving the target. This probability of success should be consistent with the participant’s tolerance for pension outcome risk. Probability of success is therefore not an objective in isolation. It is assessed together with the pension target and the amount of investment risk that is economically and institutionally acceptable.
TDC does not imply an individual investment contract or necessarily a totally customised portfolio. It provides an additional, goal-oriented layer within the collective pension arrangement and can be applied under the Dutch pension rules within both the solidarity-based pension arrangement (SPR) and flexible pension arrangement (FPR), taking account of their different investment and allocation mechanisms.
The right risk-return trade-off
The traditional risk-return framework focuses on volatility around expected return. In a standard DC approach, participants are typically assigned a risk profile that is translated into an investment policy or lifecycle.
For pension participants, however, the more relevant question is whether sufficient capital will be accumulated to achieve the desired pension outcome. The relevant risk measure is therefore not volatility alone, but the probability of failing to achieve the pension target, considered together with the magnitude and nature of the risks required to pursue that target.
Two participants of the same age can have very different financial positions. Their remaining investment horizon may be identical, while their required future return and probability of achieving the same target are materially different.
The relevant variable is therefore not age alone, but financial position relative to the target. The core logic is to link the pension target, required capital, probability of success and risk tolerance directly to investment allocation.
Designing the target defined contribution framework
The desired pension outcome provides the starting point for the investment analysis. ALM tools are required because contributions, actuarial assumptions, salary and career patterns and the economic environment all influence the pension outcome.
The pension target can be represented by a normative or notional economic pension liability: the capital required, given the relevant assumptions, to achieve the desired pension outcome. Then Monte Carlo simulations can be used to generate numerous economic, investment and participant-specific scenarios. The resulting distribution of pension outcomes can then be used to estimate the probability that the specified pension target is achieved. The probability is therefore not derived from the expected return alone. It depends on the full distribution of outcomes, including the investment horizon, contribution pattern, volatility, correlations, salary development, actuarial assumptions and other relevant model parameters.
This probability changes over time as contributions are made, investment returns are realised and the participant's financial position changes. The appropriate investment strategy will therefore also change.
The participant grid
The participant grid provides the basis for assessing participants according to their financial position rather than age alone. The primary dimensions are participation years and pension age. Additional dimensions can include accumulated capital, contributions, salary, career development, retirement age and pension target.
Figure 1 – Participant grid
Figure 1 illustrates how participants with different participation histories can have different financial positions despite having the same pension age. The grid provides the basis for calculating the capital required and the return required to reach the pension target.
The participant grid should therefore not be interpreted merely as a classification system. It is an economic mapping of the participant's position relative to the pension objective. Movement through the grid represents changes in the participant's distance-to-target.
Target annuity versus expected annuity
An important distinction exists between a target annuity and an expected annuity. If the expected return of an investment strategy equals the return required to reach the target, this does not imply a sufficiently high probability of achieving that target. The probability depends on the full distribution of outcomes, the investment horizon, contributions and other relevant assumptions.
The objective is therefore not simply to maximise expected annuity. Nor is the objective to maximise the probability of success without regard to risk. Instead, the investment strategy should provide an appropriate probability of achieving the defined pension target, consistent with the participant’s tolerance for pension outcome risk.
This distinction is central to TDC: the target defines what the participant is trying to achieve, the probability distribution describes the likelihood of achieving it, and the risk tolerance defines the amount of investment risk that can appropriately be taken to pursue that outcome.
From the participant grid to a target path
The participant grid can be extended into a dynamic target framework. The target path provides an economic reference point against which the participant's financial position can be monitored.
The following sample illustrates the mechanics of the framework. The assumptions are illustrative and are not intended as a prescribed Dutch pension target. We assume illustrative policy parameters, not regulatory requirements:
- entry age: 20;
- retirement age: 65;
- accumulation period: 45 years;
- contributions: 12% of salary.
For participants at target or less than 10 years behind, the illustrative target is 30% of salary in real terms. For participants more than ten years behind, the target is progressively reduced as the remaining horizon shortens.
Figure 2 – Model outputs under stated assumptions, not empirical estimates or guarantees.
For a participant following the target path, the annual actuarial return required to achieve the illustrative 30% real annuity target is approximately 4.36%. Under the assumptions of the sample design, this rises to approximately 6% at 44 years of missed participation.
The required actuarial return is not itself the investment recommendation. It is an intermediate economic variable. The investment allocation must be evaluated against the distribution of returns that different portfolios or collective investment strategies can generate and against the applicable risk tolerance.
The purpose of TDC is therefore to determine whether a feasible investment strategy can provide an appropriate probability of achieving the target, rather than simply to identify the return mathematically required to achieve it.
Based on the required actuarial return and the other assumptions of the model, the annual target annuity levels can then be calculated over the course of participation. Figure 3(a) shows the target annuity levels at retirement depending on the number of missed participation years. Figure 3(b) shows the target annuity path for the at-target participant. These are model outputs under the stated assumptions, not expected market returns or guarantees.
Figure 3: Target annuity and required return under stated assumptions, not expected market returns or guarantees.
Once the pension target has been defined, a probability of success can be assigned to each position in the participant grid. In the illustrative design, the probability of achieving the 30% real annuity target for a new participant entering at age 20 is 68%, increasing to 80% at retirement as the participant progresses along the target path.
These probability levels should be understood as model parameters or policy objectives within the illustrative framework, rather than as universal optimal levels. In an actual application, the appropriate probability of success would need to be determined in conjunction with the collective risk tolerance, participant characteristics and pension objective.
Having set the above scheme parameters we also model participants in terms of their salary development. Like many aspects of the design this is dependent on the population at hand and should be looked at in each specific case.
Figure 4 – Salary and Career Development
Figure 4a shows assumed starting salaries by age. In the sample, a 20-year-old participant starts with €35,000 and a 40-year-old participant with €65,000. While figure 4b shows the assumed age-dependent career development. Salary growth decreases with age and reaches zero at age 60. These assumptions are illustrative and should be calibrated to the relevant participant population in an actual application.
TDC within collective DC arrangements
TDC does not require every participant to have a separately managed investment portfolio. The framework can identify differences in participants' economic positions and translate these into differences in desired risk exposure, while implementation remains collective. The collective risk tolerance provides the boundary within which target-oriented decisions can be made.
In the SPR, implementation must remain consistent with the collective investment and allocation mechanisms of the arrangement. In the FPR, the framework can support differentiated investment profiles where appropriate. The objective is not maximum individualisation. It is to determine where differentiation creates sufficient economic value to justify additional complexity.
From age-based lifecycle to objective-based investment
TDC does not reject lifecycle investing. Age remains an important determinant of investment horizon and risk capacity. TDC adds another dimension:
What is the participant's financial position relative to the pension target?
Two participants of the same age can therefore have different required returns and probabilities of success. The framework can identify situations in which an age-based lifecycle may be insufficiently aligned with the participant's actual financial position. The purpose is not to replace lifecycle investing, but to address the fact that age alone does not provide sufficient information for the required risk allocation.
Conclusion
The transition from Defined Benefit to Defined Contribution makes the connection between the pension objective, the participant's financial position and investment policy more important. TDC creates this connection by linking these three elements:
The pension objective defines the economic reference point, the financial position shows where the participant stands relative to that objective, and the investment strategy is aligned with the probability of success and risk tolerance, within the collective constraints of the arrangement.
Pension objective → individual target → distance-to-target → probability of success → risk tolerance → investment strategy
TDC does not replace ALM, lifecycle investing, collective investment or the applicable Dutch pension framework. Instead, it adds an objective-driven layer that connects these existing components to the economic position of the participant.
The essential distinction is that TDC does not start with the question of how much investment risk a participant should have at a particular age. It starts with the pension objective and asks what financial position is required to achieve that objective, what probability of success can reasonably be pursued, and what risk tolerance is available to do so.
The resulting framework can therefore be summarised as:
Pension objective → target → financial position → distance-to-target → required return → probability distribution → probability of success → risk tolerance → investment strategy → dynamic monitoring
TDC therefore represents an evolution from an age-driven investment architecture towards an objective-driven architecture, while retaining the collective structure and ALM discipline of the modern pension arrangement. Part 2 of the position paper will focus on the decumulation phase and how to convert capital into sustainable real income while managing longevity, investment and sequence-of-returns risk.