BI Rate and Minimum Wage Are Rising, What Happens to Employee Pensions?
Two big pieces of 2026 news are moving in the same direction — but their effect on a company’s post-employment benefit obligation can point the opposite way. This S&R simulation traces why, from the employee’s perspective all the way to the CFO’s.
A note on terms: the word “pension” in the headline is used as accessible, everyday language. This article is not only about pension schemes, but about post-employment benefits as a whole — including severance pay, long-service awards, and other benefits arising from the end of an employment relationship. The technical term is Defined Benefit Obligation (DBO), and it is not the same thing as “pension money” alone.
BI Rate, SUN yield, and the minimum wage — figures that actually happened, with source & date.
DBO Rp100 bn, 7% discount rate, 8% salary growth, 8/7-year duration — model parameters, not market data.
Every %ΔDBO figure, table, and chart — calculated from (1) and (2). This is sensitivity/scenario analysis, not a forecast.
Summary of all model limitations (for readers who want to jump straight to the numbers)
- The duration approach ignores convexity — accuracy declines for Δr >150bps.
- Current service cost, past service cost, benefit payments, & demographic changes are not explicitly modelled in the main simulation — see the conceptual roll-forward in Section 12.
- The scheme is assumed unfunded — results differ for a funded scheme (see Section 10).
- Deferred tax on OCI remeasurement is modelled conceptually, not as an actual tax calculation for any specific company.
- ROE and L/E are illustrative sensitivity, not official ratios or a performance forecast — ROE uses ending-period equity, not the average.
- All sensitivity figures (breakeven ≈82bps, +14.5% per 1pp of salary, etc.) are an illustrative management scenario specific to this model’s parameters — not a universal actuarial benchmark.
- The covenant headroom in Section 18 is an illustrative simulation; the real ratio definition depends on each company’s financing agreement.
Full detail on each point is in the relevant section and the Appendix.
One Question From 2026
Interest rates are rising. The minimum wage is rising. What happens to a company’s post-employment benefit obligation?
Most people’s intuition would say: it must go up, since both are moving in a "heavier" direction. But the real answer is more interesting: not necessarily — and the direction could even be the opposite. This article traces why, starting from employee rights all the way to the impact on financial statements and CFO decisions.
Sections 01–06 build the conceptual foundation. Sections 07–13 break down the mechanics of the minimum wage, interest rates, funded/unfunded schemes, BPJS, accounting, and tax. Sections 14–20 apply all of this to 2026 conditions and its consequences for management. Full mathematical model detail is in the Appendix.
Two Important Events in 2026
Through August 2026, two developments have been moving in the same direction.
First, the BI Rate rose. From 4.75% at the end of 2025, the BI Rate rose in stages to 5.75% by mid-2026, and held at that level through July.
Source: Bank Indonesia, consecutive Board of Governors’ Meeting decisions throughout 2026. ACTUAL DATA
Second, the minimum wage rose. The government set the 2026 minimum wage increase at a national average of around 5–6%, though the amount varies by province — some below that, some approaching double digits in provinces with high economic growth.
Both events move in the same direction: up. But the question differs depending on who is asking.
For Employees
Will the minimum wage increase raise the income they receive now — and the benefits they may receive when their employment ends?
For Companies
Do the minimum wage increase and the change in interest rates automatically raise the post-employment benefit cost that must be recorded?
The answer: not necessarily. That is precisely what makes the issue interesting — and this article starts from the employee side first, before moving to the company side.
From the Employee’s Side: An Accumulating Right
Before discussing accounting, start with the most important question: what does the employee actually accumulate during the years of service?
When an employment relationship ends under conditions covered by applicable manpower regulations, a worker may be entitled to various forms of compensation — including severance pay, long-service awards, compensation for rights, separation pay, or other benefits depending on the circumstances. The type and amount of the benefit are not necessarily the same for every employee; they depend on how and when the employment relationship ends.
As the employee continues to work, the benefit attributable to that service accumulates. From the employee’s perspective, it can be viewed, in simple terms, as an accumulating retirement entitlement — almost like a “pension savings” balance built up through years of service. This is an analogy for understanding the accumulation of rights, not a statement that every benefit is held in an individual savings account.
From the company’s perspective, the same benefit represents an accumulating cost and obligation arising from services already received from the employee. The company therefore does not wait until the employee actually leaves or retires before considering the obligation.
This is technically broader than the word “pension”. The discussion covers post-employment benefits as a whole, including statutory termination benefits and other benefits arising from the end of an employment relationship. Formal pension schemes and BPJS Employment Social Security are separate arrangements and are discussed later.
Why Does Tenure Matter?
The logic is simple: the longer an employee has served the company, the larger the portion of benefits that has conceptually been earned through that service.
That’s why a company doesn’t wait until the benefit is actually paid to start recognising the obligation. This obligation accumulates every year, in step with an employee’s tenure — it doesn’t suddenly appear in the year an employee leaves.
A note on terms: this section deliberately avoids the phrase "pension expense". The term used consistently throughout is employee benefits / post-employment benefits — because, again, its scope is broader than pensions alone.
Why Does a Company Need Probability?
Why isn’t it enough for a company to simply multiply the number of employees by the maximum benefit?
Because a company doesn’t know for certain how each employee’s working relationship will end. Different possibilities produce different benefits:
• Reaching retirement age
• Dying before retirement age
• Becoming disabled
• Resigning
• Being laid off
• Other conditions under applicable regulations
Because the entitlement and the size of the benefit depend on which condition occurs, actuarial work needs to weigh the probability of each condition happening. Conceptually:
Expected Benefit = Benefit × Probability of Occurrence
An employee is entitled to a hypothetical Rp500 million benefit if they reach retirement age at the company. But that outcome isn’t 100% certain:
• Probability of staying until retirement ≈ 55% → full Rp500 million benefit
• Probability of resigning before retirement ≈ 35% → a much smaller benefit (per resignation terms)
• Probability of death/disability before retirement ≈ 10% → a benefit under a different formula
The percentages above are purely illustrative — real figures come from mortality tables, disability data, and the company’s own turnover experience. The point: this employee’s expected benefit is not the full Rp500 million, but a weighted combination of all those possibilities.
This is why the DBO is not the same as the maximum nominal benefit multiplied by headcount. The DBO is an estimate of the present value of the benefit obligation that is probabilistically expected to be paid — not the maximum amount the company must pay.
How Is the DBO Formed?
Before getting into the formula, understand the conceptual flow first:
This also reveals a bigger logical chain: labour law/regulation determines an employee’s entitlement; actuarial work translates that into an expected obligation through probability, projection, and discounting; and PSAK 219 sets out how that obligation is measured and presented in the financial statements. Three different things, three different stages.
Projected Unit Credit, in Plain Language
Once the mechanism above is understood, the formal method can be explained briefly: Projected Unit Credit allocates the projected benefit across an employee’s years of service, then measures the present value of the portion "earned" (accrued) as of the reporting date. That’s enough to grasp the essence — further technical detail is in Appendix A.1.
PSAK 219 sets out the principles for recognising, measuring, presenting, and disclosing employee benefits, including measuring the defined benefit obligation using a relevant actuarial method. PSAK does not provide one simple formula you plug numbers into — an actuary performs the valuation based on relevant data and assumptions.
Data and Assumptions: Where Do the DBO Numbers Come From?
Actual data describes what has already happened; assumptions are used to estimate what might happen in the future. Distinguishing the two matters — it’s why the 2026 minimum wage is a fact, while the 2026–2035 salary growth assumption is a projection.
| Category | Examples |
|---|---|
| Company internal data | Employee age, tenure, salary, employment status, turnover history, benefit payment history, workforce structure |
| External data | Manpower regulations, minimum wage, market yield, economic data, mortality tables/actuarial basis |
| Assumptions | Salary escalation, turnover, mortality, disability, retirement age, discount rate |
Three Groups of Assumptions
Instead of splitting assumptions into a binary "controllable" vs "uncontrollable" — a distinction that’s too harsh and not entirely accurate — it’s more appropriate to use three groups:
Salary increase policy, workforce planning, retention, recruitment, retirement policy. Can be reviewed and adjusted by management using internal data.
Statutory minimum wage, market yield, monetary policy, regulatory change. Set by market conditions and government policy.
Mortality, disability, turnover, retirement behaviour. A probability basis reviewed by actuaries from historical data and reference tables.
Important note: "management-influenced" does not mean fully controllable by management. Salary policy is still constrained by labour market conditions and industry competitiveness; retirement age is still constrained by regulation. This is a framework for understanding who reviews an assumption, not a claim of full control.
Minimum Wage Rises: Does the DBO Definitely Rise?
Does a minimum wage increase always raise the DBO?
Not automatically. The impact depends on whether the minimum wage increase actually raises that company’s employee salaries, and whether the change triggers a revision of the long-term salary growth assumption.
Company A — labour-intensive manufacturing
Many employees earn near the minimum wage → a minimum wage increase has a larger effect on the DBO.
Company B — professional services
Salaries are well above the minimum wage → the impact can be small or non-existent.
Actual salary increase (salary genuinely rising this year) is different from salary escalation assumption (the long-term expectation used to project benefits). The minimum wage can be one input considered when revising the latter — but not automatically.
An Indirect Effect: Wage Compression
A minimum wage increase can also have an indirect effect that’s rarely discussed. When the minimum wage rises, the pay gap between minimum-wage workers and the level above them — supervisors, for instance — narrows. Companies often respond by adjusting pay for levels above minimum wage too, to keep the salary structure proportional.
An illustrative example: an operator gets a 6% raise following the minimum wage, but a supervisor can’t realistically stay at the same nominal gap. The company might raise the supervisor 5%, middle management 3%, and so on — the minimum wage effect ripples through the entire pay structure, not just the workers who receive it directly.
As an illustration only: in a model with a 7-year salary duration, a 1 percentage point increase in the long-term salary growth assumption pushes the DBO up net by around 14.5%. This is an illustrative management scenario, not an actuarial benchmark — the figure depends heavily on the duration parameter used, and can be very different from one company to the next. The layered logic matters more than the number: minimum wage rises → not every employee’s salary necessarily rises → even if it does, that doesn’t necessarily change the salary escalation assumption → only then might the DBO change.
The BI Rate Rises: Why Can the DBO Actually Fall?
Up to this point, the link between the minimum wage and post-employment benefits is relatively easy to grasp: if the salary underlying the benefit rises, the estimated future benefit can rise too.
But what about the BI Rate? Isn’t the interest rate mainly about loans, deposits, and a company’s finance costs?
It turns out it’s not that simple. Because post-employment benefits are paid in the future, while the DBO must be valued at present value as of the reporting date — the higher the discount rate used, the smaller the present value recorded today for the exact same future obligation.
Discount rate rises → Present value falls → DBO falls
Discount rate falls → Present value rises → DBO rises
But there is one more mechanism working alongside it, and it’s often missed: the DBO doesn’t just sit still — it keeps growing in value as time passes, even with no change in assumptions at all. This effect is called interest accretion (or discount unwinding), and it is always positive.
Opening DBO: Rp100 bn
+ Interest accretion (7% discount rate × Rp100 bn): +Rp7 bn
− Remeasurement (discount rate up 100 bps, 8-year duration): −Rp8.56 bn
= Closing DBO: Rp98.44 bn — a net decline of 1.56%, not 8%
The gold bar (interest accretion) always adds. The green bar (remeasurement) subtracts because rates rose. The net result is far lighter than remeasurement alone might suggest. The full scenario range is in Section 16.
Model note: the calculation above isolates two mechanisms (interest accretion and remeasurement) for sensitivity purposes — it is not a full actuarial reconstruction of the DBO roll-forward (see the full roll-forward in Section 12).
The BI Rate Is Not the PSAK 219 Discount Rate
An important clarification: readers should not conclude that "BI Rate up 100 bps = PSAK 219 discount rate up 100 bps". That would be wrong.
PSAK 219 requires a discount rate derived from high-quality corporate bond yields, or, where no deep enough market exists for such bonds, government bond (SUN) yields. The BI Rate influences that yield, but through a transmission mechanism — not directly:
So the BI Rate is an economic driver, not the discount rate itself. As of 5 August 2026, the 10-year SUN yield stood at 7.31% (Bloomberg Technoz) — well above the 5.75% BI Rate, and the relationship between the two is not a proportional 1:1. Empirical evidence and full transmission detail are in Appendix A.4.
S&R Sensitivity & Scenario Analysis
Before the numbers: it’s important to distinguish three terms that are often conflated. Sensitivity analysis answers "what happens if one variable changes?" Scenario analysis answers "what happens under a combination of assumptions?" A forecast answers "what do we actually expect to happen?" — this section is S&R sensitivity and scenario analysis, not a forecast, because the SUN yield and salary escalation here are illustrative assumptions, not official predictions.
16.1 — Sensitivity to Interest Rates
The table below traces various rate-increase/decrease scenarios, holding the salary growth assumption constant.
| Δ Rate (bps) | New Discount Rate | Interest Accretion | Remeasurement | Closing DBO | Net %Δ DBO |
|---|
Yellow row = base scenario (no change) — the DBO still grows +7% purely from interest accretion. S&R CALCULATION
The pattern: the "Net %ΔDBO" column is far flatter than the "Remeasurement" column alone. Interest accretion keeps offsetting part of the decline, up to a point where the rate increase is large enough to overcome it.
Net %ΔDBO vs. pure remeasurement
Gold line = net result (after interest accretion) · dashed line = remeasurement alone
The DBO only shrinks net once rates rise more than ≈82 bps within a year — this threshold is not a fixed number; it depends on the company’s liability duration (8 years in this model). See Appendix A.3 for how to adjust it to your own company’s profile.
Limitation note: this duration approach is a first-order approximation — for large interest rate changes (>150 bps), its accuracy declines because it ignores convexity (a non-linear effect). Mathematical detail is in Appendix A.2.
16.2 — Sensitivity to the Salary Assumption
| Δ Salary Assumption (pp) | New Assumption | Closing DBO | Net %Δ DBO |
|---|
In S&R’s illustrative model with a 7-year salary duration. The +14.5% figure for a +1 pp assumption increase is not a general estimate — the result depends entirely on the assumed liability profile. A company with a different salary duration would get a different figure. S&R CALCULATION
Company A, which revises its salary assumption up 1pp, moves well away from the base line. Company B, which revises nothing, stays at +7% (pure interest accretion).
16.3 — Combining Interest Rate & Salary
In the real world, interest rates and the salary assumption rarely stand still one at a time. They pull the DBO in opposite directions — rising rates push it down, rising salary/minimum wage push it up. The matrix below maps their combined, simultaneous effect.
Rows = change in discount rate (bps) · Columns = change in salary assumption (pp). S&R CALCULATION
Most cells are positive — a sign that interest accretion is strong enough to offset many combinations of rising rates and salaries at once. The DBO only shrinks net in the combinations toward the bottom-right of the matrix, when rates rise sharply while salaries are not revised.
Which scenarios ease or worsen the DBO most?
Only the scenario with rates sharply up and a wage-neutral outcome ends with a net DBO decline. Other scenarios — including those combining a moderate minimum wage increase — still end with the DBO growing, because interest accretion adds weight in every scenario.
The DBO Is Not Always the Same as the Net Liability
So far this article has treated the DBO as if it were the full liability recorded on the balance sheet. That’s true for most companies in Indonesia — but not for all.
Unfunded
No qualifying plan assets are set aside separately. The net defined benefit liability is close to the DBO. Main focus: liability risk alone. This is used as the basis for the simulation throughout this article.
Funded
Plan assets exist. Net defined benefit liability = DBO − Fair Value of Plan Assets. What matters is not the DBO alone, but the net position.
A Funded Plan Has a "Mirror Image" of Interest Rate Risk
For a funded scheme, a higher discount rate can indeed lower the DBO — but the value of plan assets can also change, depending on the portfolio composition. A company with a funded scheme faces liability risk plus asset/investment risk at once, not liability risk alone. A rising discount rate can lower the DBO, but a change in plan asset value can offset it or amplify the change in net position — management needs to look at the asset-liability interaction, not the DBO in isolation.
For professional readers: in a funded scheme, recognition of a net defined benefit asset is also subject to the relevant asset ceiling provisions — not every surplus of plan assets over the DBO can automatically be recognised as an asset without limit. Further detail is in Appendix A.7.
What About BPJS Ketenagakerjaan?
BPJS Ketenagakerjaan (Indonesia’s workers’ social security scheme) is often assumed to "already cover" a company’s post-employment benefit obligation. That’s not entirely accurate — BPJS should not be treated as synonymous with the DBO. All three are different layers:
Does BPJS eliminate a company’s post-employment benefit obligation?
Not automatically. An employee can have BPJS coverage, and at the same time the company can still have an employee benefit obligation under manpower regulations or its own programme. BPJS is not an automatic replacement for the employer’s obligation — the treatment must be assessed based on the type of benefit and the characteristics of each programme, not assumed to fully overlap without examining the scheme and its legal basis.
Same Event, Different Consequences
A single economic event can affect every layer differently, all at once:
| Event | Employee | Payroll Cost | BPJS Cost | DBO |
|---|---|---|---|---|
| Minimum wage ↑ | may ↑ | ↑ | may ↑ | may ↑ |
| BI Rate ↑ | not automatically ↑ | not automatically ↑ | not automatically ↑ | may ↓ |
| Salary policy ↑ | ↑ | ↑ | may ↑ | may ↑ |
| Discount rate ↑ | benefit formula unchanged | — | — | may ↓ |
This table is a conceptual summary of every mechanism discussed in Sections 02–10 — one event, five consequences that can point in different directions.
How Does a Change in DBO Enter the Financial Statements?
Every %ΔDBO figure above eventually lands in the financial statements — but not all in the same place, and not only through the interest rate mechanism.
DBO Roll-Forward (Conceptual)
The S&R sensitivity model in this article deliberately isolates interest accretion and remeasurement. But in a full actuarial valuation, the DBO moves through more components:
The S&R model in this article only isolates the "net interest" and "remeasurement" lines for sensitivity purposes — not a full roll-forward reconstruction. Current service cost, past service cost, curtailment, and settlement are discussed further in Appendix A.8.
Remeasurement itself doesn’t only come from changes in the discount rate and salary assumption. It can also come from changes in other financial assumptions, changes in demographic assumptions (mortality, disability, turnover), or an experience adjustment — the difference between actual experience and the assumption previously used (for example, actual turnover differing from what was expected). This is what connects the probability concept in Section 04 to the accounting consequences here.
| Component | Recognised In |
|---|---|
| Current service cost | P&L |
| Net interest | P&L |
| Remeasurement (assumption change) | OCI |
OCI doesn’t mean "unimportant" — its impact bypasses current-period profit or loss, but it still affects equity and financial ratios. This simulation assumes an unfunded scheme (no separate plan assets, as is commonly found at many Indonesian companies); for a funded scheme, net interest is calculated from the net liability, not the gross DBO (see Section 10). Deferred tax effects on OCI remeasurement are also not modelled numerically — in practice, a DBO change also moves deferred tax assets/liabilities, so the L/E and ROE figures below likely overstate the net impact on equity compared with the post-tax position (see Section 13).
⚠ When ROE Rises Not Because Profitability Improved
DBO rises → Liabilities rise → Equity falls (via OCI)
→ L/E rises but net income does not automatically fall
→ ROE can rise because its denominator (equity) shrinks — not because performance improved
A mechanical illustration: net income is held constant to isolate OCI’s effect on equity. This is "illustrative ROE sensitivity", not an official ROE or performance forecast — it is calculated from ending-period equity, not average equity (opening+closing)/2 as in standard ratio analysis practice, which is used to avoid distortion from equity transactions late in the year. In actual reporting, service cost and net interest can also differ across scenarios. S&R CALCULATION
The scenario with the worst L/E has the best ROE. Looking at a rising ROE without checking L/E can lead to a conclusion opposite to the real picture.
What About Taxes?
A change in DBO also has tax consequences, even if indirect. There are three "tax moments" that need to be separated, because their timing differs:
| Moment | Explanation |
|---|---|
| Accounting recognition | The DBO is recognised under PSAK 219, at the point the benefit is "earned" for accounting purposes. |
| Tax treatment | Expense recognition for tax purposes follows relevant tax regulations and timing — usually different from the accounting timing. |
| Payment / withholding | When the benefit is actually paid, tax consequences arise for the recipient and a withholding obligation follows depending on the type of payment. |
Accounting timing ≠ Tax timing ≠ Cash payment timing.
For specific tax figures and treatment, always verify against the latest tax regulations and consult your tax team — this article deliberately avoids giving specific tax treatment examples to avoid overly broad generalisation.
Deferred Tax, Conceptually
As a result, the after-tax impact can differ from the gross change in DBO. This is far more useful to a CFO than simply noting "there is a tax effect" — because it means the illustrative L/E and ROE figures in Section 12 need to be read with the reminder that they may not reflect the post-tax position.
A Rp10 billion increase in DBO does not mean the company must pay Rp10 billion today. Conversely, a Rp10 billion decrease in DBO also does not mean the company receives Rp10 billion in cash.
Distinguish three things: accounting obligation (the number recorded on the balance sheet) vs future cash requirement (when and how much will actually be paid) vs, for a funded scheme, funding requirement (how much needs to be contributed to plan assets).
What Has Happened So Far, Through August 2026?
All the frameworks above (Sections 02–06) hold for any duration or discount rate. Let’s now apply them to actual market figures as of August 2026.
The scenarios in this section only hold for market conditions as of the dates shown — market data moves quickly. To update: take the current 10-year SUN yield (Bloomberg/Refinitiv/IBPA), plug it into the Appendix A.2 formula, and adjust the duration using your own company’s Notes to the Financial Statements — this calculation can be replicated at any time.
ACTUAL DATAThe 10-year SUN yield rose from ≈6.1% in early 2026 to 7.31% on 5 August 2026. The BI Rate is not the discount rate PSAK 219 calls for — the standard requires a high-quality corporate bond yield, or, where none is available, a government bond (SUN) yield. The BI Rate is only one factor influencing that yield, and the relationship is not a proportional 1:1 (transmission detail & evidence of non-linearity are in the Appendix).
What Might Happen Between Now and December?
Actual data only runs through August, while financial reporting ends in December. The timeline:
For year-end 2026: Bank Danamon (Irman Faiz, 23 Jul 2026) projects a 6.25% BI Rate; PT SMF (Martin Daniel Siyaranamual, 22 Jul 2026) projects 6.25–6.75%; LPEM UI projects the BI Rate flat at 5.75%. A bond analyst (quoted by Kontan/MomsMoney, 14 Jul 2026) projects the 10-year SUN yield at 6.8–7.5%. These targets belong to the sources named, not an S&R projection — and market data moves quickly; these projections can change after each publication date.
S&R SCENARIO CALCULATION| Scenario | Target SUN Yield | Salary Assumption Revision | Net %ΔDBO Full-Year |
|---|
From the end-2025 base (≈6.15%) to the end-2026 target, already including a full year of interest accretion. S&R CALCULATION — this is scenario/sensitivity analysis, not an official S&R forecast.
What Can Still Change Before December?
The figures above are August sensitivity — not the December DBO. What can still change before year-end close includes:
• Further BI Rate and market yield moves
• Internal salary adjustments
• Workforce changes: resignations, layoffs, restructuring
• Acquisitions or mergers
• Benefit payments that have already occurred
• Other actuarial assumption changes
August sensitivity is not December DBO.
From DBO to Projected Financial Statements
This is what connects actuarial work to CFO decisions. After the DBO simulation, the key question is: so what?
DBO rises → liabilities rise → equity can fall via OCI (with a relevant tax effect, see Section 13) → the leverage ratio changes → projected FS changes → covenant headroom changes → management action may be needed. Each stage has already been discussed separately in earlier sections — this is where it all connects into one decision chain.
What About Debt Covenants?
A change in DBO can affect certain financial ratios — but it’s not correct to say "a rising DBO must breach a covenant". It all depends on the ratio and covenant definitions in the relevant financing agreement.
Potentially relevant ratios: liabilities/equity, debt/equity, leverage ratio, net debt/EBITDA, minimum net worth, interest coverage. Not all of them are affected the same way, and some covenants even have defined adjustments to the accounting figures (for example, excluding the impact of OCI remeasurement). Every covenant must be read according to its own contractual definition.
Illustration: Covenant Headroom
As an illustration (not a real figure for any company) — suppose a covenant caps the liabilities-to-equity ratio at a maximum of 1.00x. Here is the projected headroom across the five scenarios from Section 16:
| Scenario | Projected L/E | Covenant Limit | Headroom |
|---|
Headroom = Covenant Limit − Projected Ratio. L/E figures from the combined scenarios in Section 16. S&R CALCULATION — ILLUSTRATIVE
With this framework, this article is no longer just "PSAK 219 explained" — it shows how a single actuarial assumption can become a financing issue. That is what makes sensitivity analysis relevant to a CFO, not just to the accounting team.
What Should Management Do Before Year-End Close?
One thing needs to be clarified first: management does not set the discount rate. A CFO cannot and should not "adjust" the discount rate to make the DBO look good. What management can do is manage the consequences — not the variable itself.
"Management does not manage the discount rate; management manages the consequences."
Pre-Close Checklist
- Update the employee census data.
- Review the salary structure after the minimum wage increase.
- Review the long-term salary escalation assumption.
- Review the discount rate basis / current market yield.
- Review demographic and turnover assumptions.
- Update the actuarial valuation/sensitivity.
- Assess the funded vs unfunded position (Section 10).
- Reconcile BPJS data and employer contributions (Section 11).
- Assess the accounting and tax impact (Sections 12–13).
- Update the projected financial statements.
- Calculate covenant headroom (Section 18).
- Identify any management action that may be needed before close.
If management only sees the DBO figure once the December actuarial report is complete, there is very little room left to act. Sensitivity analysis in August/September gives time to assess the impact on payroll, funding, tax, projected FS, and covenants — before year-end close, not after.
Conclusion
Summarising the entire article into the two perspectives that have framed it from the start.
For Employees
Does a rising interest rate mean my post-employment rights are reduced?
No. A change in the discount rate mainly changes the present value of the obligation the company measures and records — it does not change the nominal benefit formula employees are entitled to under manpower regulations. DBO falls because the discount rate rises ≠ employee entitlements fall. Conversely, a minimum wage increase also doesn’t automatically mean post-employment benefits rise too — that depends on the salary structure of the company they work for and the long-term assumption used.
For CFOs, Finance, Accounting, and Tax
The question is different: how much liability must be reported? What’s the impact on P&L and OCI? What’s the equity position? What’s the tax effect? Are covenants still safe? What’s the cash requirement? What can still change before December? What matters isn’t just today’s DBO, but how the DBO is projected through year-end, and how that affects the company’s projected financial statements, debt covenants, equity, financial ratios, and tax position.
- A rising interest rate does not automatically make the DBO fall net — interest accretion works against remeasurement, and often offsets most of the decline.
- A rising minimum wage does not automatically make the DBO rise — it depends on the company’s salary structure and whether the increase triggers a long-term assumption revision, including through the wage compression effect.
- OCI remeasurement does not affect current-period profit or loss — but it still affects equity, L/E, and can make the illustrative ROE rise without any improvement in performance.
- The numbers in this article depend on the duration, discount rate, and funded/unfunded scheme assumptions used — use data from your own company’s Notes to the Financial Statements for a relevant result (see the Appendix).
Myth vs Fact
"Interest rates rise → the DBO must fall."
Not always. Interest accretion keeps running and can offset most of the decline.
"Minimum wage rises → every company’s DBO rises."
No. It depends on the salary structure and whether the increase changes actual employee pay.
"Large remeasurement → profit must fall."
No. Remeasurement goes to OCI, not P&L.
"DBO falls → the company must be healthier."
Not necessarily. Financial ratios must be read alongside other components, including the post-tax position.
What should management watch for?
Not just "what will interest rates be next year?" — but the direction of the discount rate, the duration of the obligation, salary growth policy, the workforce’s demographic profile, the funded/unfunded position, and the tax & covenant consequences, all at once.
The DBO is not just an actuarial number. It is a mirror of a company’s promise to its workforce, financial market conditions, and the company’s own workforce structure — and understanding it requires the lens of employee benefits, actuarial science, accounting, tax, and financial planning all at once.
Appendix: The Full S&R Model
This section contains all the mathematical detail, verification, and technical notes underlying Sections 00–20 — for readers who want to check, replicate, or adapt the model to their own company.
A.1 — Model Parameters
Both duration figures are illustrative assumptions for a workforce with an average age of ≈35 and ≈25 remaining years of service. If your company’s profile differs, the sensitivity results scale proportionally — see A.3.
More formally: Projected Unit Credit calculates the present value of the benefit "earned" (accrued) as of the reporting date, by allocating the total projected benefit proportionally across each year of an employee’s service (usually a straight-line method per year of service, unless the benefit formula states otherwise). The S&R sensitivity model in this article does not perform an individual per-employee, per-year-of-service allocation — it works on an aggregate DBO basis to isolate the duration effect.
A.2 — Base Formula & Sensitivity Threshold
%ΔDBO ≈ −Dr × Δr (interest rate effect)
%ΔDBO ≈ +Dg × Δg (salary growth effect)
Closing DBO = Opening DBO + Interest Accretion ± Remeasurement. From this, a threshold can be derived: the DBO keeps growing net as long as Duration × Rate increase < Opening discount rate ÷ (1+Opening discount rate), or as a unitless ratio R = (Duration×Δr)/(r₀/(1+r₀)); R<1 the DBO grows, R>1 the DBO shrinks.
Within this illustrative model’s framework, that relationship holds consistently across every parameter combination tested — a mathematical property of this first-order duration approximation model, not a claim that this linear relationship is exact in an actual actuarial valuation.
Formula verification on 9 extreme combinations (of 75 tested)
| Discount Rate | Duration | ΔRate | Ratio R | Net %ΔDBO | Prediction |
|---|
Combinations were deliberately chosen far from the 7%/8-year base to show the relationship stays consistent.
A.3 — Sensitivity to Duration (How to Adapt It to Your Company)
Take your company’s liability duration and discount rate from the Notes to the Financial Statements, employee benefits section. The table below shows how the breakeven point shifts with duration.
| Duration | Illustrative Profile | Breakeven Point (bps) | Net %ΔDBO @+100bps |
|---|
A.4 — Why Is the BI Rate Mentioned, When It Isn’t What’s Used?
PSAK 219 requires a discount rate derived from high-quality corporate bond yields, or, where none is available, government bond (SUN) yields — not the BI Rate. The BI Rate influences the SUN yield through three channels: expectations (long yields = the average expected path of short rates), competition for funds (rates up → short-term instruments become more attractive → long bonds demand compensation), and independent risk (the rupiah, capital flows, fiscal conditions, US Treasuries).
Evidence the relationship is not linear (2026):
1. The SUN yield curve briefly inverted (2 Jun 2026): the 1-year yield (7.10%) exceeded the 10-year yield (6.69%).
2. BI raised rates 25 bps (19 Jun 2026), yet the 10-year SUN yield moved only +6.6 bps that day.
3. Across 2026: BI Rate +100 bps, 10-year SUN yield +116–122 bps — same direction, different magnitude.
The 10-year SUN yield is usually higher than the BI Rate because of the term premium, sovereign risk premium, long-term inflation expectations, and liquidity premium. The current gap (5 Aug 2026): 7.31% − 5.75% = +156 bps.
Pass-through table: BI Rate × Beta → Discount Rate → DBO
| ΔBI Rate | Beta | ΔDiscount Rate | Net %ΔDBO |
|---|
Empirical β for 2026 ≈1.2x (marked row): BI Rate +100bps, SUN yield +116–122bps across Jan–Aug 2026.
A.5 — Effect on Employees: Does Severance Pay Shrink?
7% discount rate → present value: Rp76,252,394
8% discount rate → present value: Rp69,479,023 (−8.88%)
The Rp150 million nominal promise does not change.
What moves as the discount rate rises is the reserve the company books today — not the size of the promise. The nominal severance pay is calculated from the Manpower Law / Government Regulation 35/2021 formula (a multiplier × last salary × years of service), which contains no discount-rate variable. "The DBO fell because rates rose" is balance-sheet news for the company, not news that an employee’s severance pay has shrunk. What matters more to employees: the company’s liquidity when the obligation falls due, since the scheme at many Indonesian companies is unfunded.
A.6 — Model Limitations
- Current service cost, benefit payments, and demographic changes are not modelled — the closing DBO here is purely the result of interest accretion + remeasurement. For fast-growing companies (startups/major expansion), current service cost can add to the DBO materially — potentially far more than the remeasurement effect discussed in this article. Conversely, for a company downsizing its workforce (layoffs/outsourcing), actual benefit payments can reduce the DBO significantly. Neither effect is captured by the model.
- Convexity (non-linearity) is ignored — for Δr >150bps, the accuracy of the duration approach declines.
- The scheme is assumed unfunded — for a funded scheme, PSAK 219 net interest is calculated from the net liability (DBO less plan assets), which can differ materially from this model (see A.7).
- Deferred tax effects on OCI remeasurement are modelled conceptually only (Section 13), not calculated numerically in the main simulation.
- The illustrative ROE uses constant net income & ending-period equity (not the average) — a simplification to isolate the OCI effect, not a standard performance-analysis methodology.
- Workforce restructuring (M&A, mass layoffs, outsourcing) can change the DBO by far more than a 50–100 bps move in the discount rate — not modelled here because it is highly specific to each transaction/company.
A.7 — Funded Plan: Asset Ceiling
For a funded scheme, recognition of a net defined benefit asset is also subject to the relevant asset ceiling provisions — not every surplus of plan assets over the DBO can automatically be recognised as an asset without limit in the financial statements. This provision caps recognition of the surplus at the present value of economic benefits available to the company, for example in the form of refunds or reductions in future contributions. Asset ceiling calculation detail is beyond this article’s scope — relevant mainly to companies with a funded pension scheme in a surplus position.
A.8 — Past Service Cost, Curtailment, Settlement
A change to an employee benefit programme, a significant reduction in the workforce covered by the programme (curtailment), or a partial/full settlement of the obligation can produce accounting consequences different from the ordinary remeasurement mechanism discussed in this article. Past service cost — for example from a change in the benefit formula — is generally recognised immediately in P&L, not through OCI. This is especially relevant for companies undergoing restructuring, mass layoffs, or a change in their pension programme (see Sections 15 & 19).
A.9 — Data Source Architecture
A summary of where each type of data/assumption should come from — useful as a quick reference for HR, finance, and actuarial teams preparing valuation data:
| Information | Primary Source |
|---|---|
| Employee age | HRIS / employee master |
| Salary | Payroll |
| Tenure | HRIS |
| Turnover | HR records |
| Benefit formula | Regulation / employment agreement |
| Salary escalation | Management policy + historical data |
| Mortality | Actuarial basis |
| Discount rate | Market yield |
| Plan assets | Custodian / investment records |
| BPJS | BPJS records |
| Tax treatment | Tax regulations |
| DBO | Actuarial valuation |
A.10 — What Auditors & Actuaries Review
For professional readers who want to know the scope of review before a DBO figure enters the financial statements:
- Data — employee census, payroll, tenure, age, termination records.
- Legal — employment agreements, collective labour agreements, company regulations, applicable law.
- Assumptions — discount rate, salary escalation, turnover, mortality, retirement age.
- Plan assets (if any) — existence, valuation, investment returns.
- Accounting — P&L, OCI, balance sheet, disclosures.
- Tax — temporary differences, deductibility, withholding obligations.