Product Specification#
Status: Draft, 2026-08-29 (research access date 2026-08-29).
Scope note. This is a standardized composite specification assembled for reference liability cash-flow modelling of a German Pflegerentenversicherung — the individual, privately written, single-life long-term-care annuity sold by a Lebensversicherer, which pays a monthly Pflegerente for as long as the insured holds a contractual Pflegegrad (statutory degree of care need). It describes no single insurer’s product, and it must not be read as one.
How this composite differs from its frlib counterpart.
frlib/products/temporaire_deces/product-spec.md is a composite of eight retrieved carriers, seven
of whose contracts were read in full. This one was drafted with nothing retrieved at all. Direct
HTTP egress from the build environment was blocked by an organisation network policy, and the
session’s WebSearch budget was already exhausted when work on this product began, so there was
neither a retrieval channel nor a search channel: the composite below was assembled from document
classes that exist and are the right kind of document for this product, and from the mechanics of
German insurance law and actuarial practice as the authoring model knew them, under the discipline
house rule 3 imposes for exactly that case.
That is how it was drafted, and not where it now stands. The policy has since been lifted and the
citations re-verified against the primary documents on 2026-08-30. Every statute cited here was read
as canonical XML from gesetze-im-internet with its amendment status (Stand) recorded; one carrier’s
complete Bedingungswerk and Produktbeschreibung were read as PDFs, and so were the PKV
Musterbedingungen for both the compulsory and the top-up cover, the DAV’s own derivation and
Pflegegrad reports, and the Destatis, vdek, GDV and Assekurata material. Twenty-seven of this
product’s thirty-six source entries now read Retrieved: yes, five read no, and four are partial;
sources.md names each and says what stopped it. What has not changed is the shape of this file:
one contract is not eight, so the variation table below remains a market-range reconstruction rather
than a survey, and every level in it is still a standardization. What has changed is that the
re-verification corrected claims the drafted text got wrong, each marked where it occurs.
Read the Retrieved line before relying on a tag: where it says yes the document was opened and
the passage read; where it does not, [R2] beside a statement about § 15 SGB XI names the instrument
the statement should be checked against and asserts nothing more — a pointer, not a certificate.
Facts carrying a source tag — [S#] (primary product document classes: Musterbedingungen, AVB,
Produktinformationsblatt, Basisinformationsblatt, Verbraucherinformation, Tarifblatt) and
[R#] (product-specific regulatory and actuarial references), both numbered per
_research/pflegerentenversicherung.md and resolved in sources.md (numbering frozen, never
renumbered), and [REG-R#] (the cross-product reference library, whose own R-numbering is distinct and
also frozen) — name the instrument the claim belongs to. Values marked std are standardizations
introduced for the reference implementation; each carries a numbered footnote giving the rationale
and, where the research file recorded one, the observed or argued range. Claims no source could
corroborate are flagged unverified, and on this product that is most of the specific numbers.
Out of scope, and said so where it matters. The soziale Pflegeversicherung of SGB XI R1 and
the private Pflegepflichtversicherung of § 23 SGB XI R7 are the compulsory first layer and are
described, not modelled. Pflegetagegeldversicherung and Pflegekostenversicherung are written as
private Krankenversicherung and are the contrast documents of this file, never its subject [S2].
Pflege-Bahr, the subsidised cover of § 127 SGB XI, is confined by statute to a tariff calculated
under § 146 Abs. 1 Nrn. 1 and 2 VAG — health business, with an Alterungsrückstellung — and may
carry no benefit beyond the graded Geldleistung R8 R12: a Pflegerentenversicherung cannot
be a geförderter Tarif, and the Zulage is not modelled. (The statute nowhere uses the word
Pflegetagegeld, as this note previously said it did.) Betriebliche Altersversorgung, Gruppenversicherung, Sterbegeldversicherung and
institutional risk transfer are outside the delib library entirely. The neighbouring biometric
product is products/berufsunfaehigkeit/ (BU_DE_S), which shares this product’s chassis, its
waiver of premium and its multi-state modelling problem, and differs in trigger, in duration and in
the age at which the risk bites.
Product overview and market role#
The three layers of German long-term-care funding#
German long-term care is funded in three layers, and a Pflegerentenversicherung is the third. The first is compulsory statutory cover — the soziale Pflegeversicherung (SPV) of SGB XI R1, or the private Pflegepflichtversicherung (PPV) of § 23 SGB XI for the privately health-insured R7; membership follows health insurance, so the layer is universal, and it is a Teilleistungssystem by design, paying defined amounts per Pflegegrad rather than the cost of care, with the residue falling on the insured person R1. That constitutive choice, made in 1994 and never reversed, is why the third layer exists as a market at all. The second layer is the insured person’s own resources; the third is voluntary private top-up or, failing that, means-tested Hilfe zur Pflege under §§ 61–66 SGB XII R24. The private product is sized against the gap the first layer leaves, and its trigger is defined by reference to the first layer [S1] [S4] — so the trigger is exogenous to the insurer, and the benefit is a Summenversicherung payable irrespective of what care actually costs.
The statutory first layer, quantified#
All amounts per calendar month, read from the consolidated SGB XI on 2026-08-30 (Stand: zuletzt geändert durch Art. 2c G v. 24.7.2026). They are the amounts § 30 SGB XI uprated by 4,5 % on 1 January 2025 and they were still in force in 2026: § 30 Abs. 1 schedules the next rise for 1 January 2028, so nothing changed on 1 January 2026 and research gap 8 is closed.
Pflegegrad |
Pflegegeld (cash, informal care) |
Pflegesachleistung (benefit in kind) |
SPV contribution, vollstationär |
|---|---|---|---|
1 |
none |
none |
131.00 EUR (flat, not graded) |
2 |
347.00 EUR |
796.00 EUR |
805.00 EUR |
3 |
599.00 EUR |
1,497.00 EUR |
1,319.00 EUR |
4 |
800.00 EUR |
1,859.00 EUR |
1,855.00 EUR |
5 |
990.00 EUR |
2,299.00 EUR |
2,096.00 EUR |
Basis: home care § 36 Abs. 3 and § 37 Abs. 1 R3; residential care § 43 Abs. 2 and Abs. 3 R4. The grade-1 figure was previously printed as 125,00 EUR; § 43 Abs. 3 reads “einen Zuschuss in Höhe von 131 Euro monatlich”.
Three readings drive the private product’s design. Pflegegrad 1 is, for cash purposes, uninsured by the state — §§ 36 and 37 both open “Pflegebedürftige der Pflegegrade 2 bis 5”, leaving 131 € towards a Pflegeheim and, for care at home, an earmarked Entlastungsbetrag of “bis zu 131 Euro monatlich” under § 45b Abs. 1, and nothing else R3 R4 R5 — which is why most private Leistungsstaffeln, and delib’s, also pay nothing at grade 1. The Pflegegeld runs between 40 % and 44 % of the corresponding Sachleistung — 43,6 / 40,0 / 43,0 / 43,1 % at grades 2 to 5 R3; the round “about 44 %” this document used to print is right at three grades of four and wrong at grade 3. That ratio is why roughly five in six Pflegebedürftige are cared for at home — 85,9 % at end-2023, against 14,1 % in vollstationäre care R18. And the residential contribution at grade 5 (2 096 €) is lower than the home Sachleistung at grade 5 (2 299 €) R3 R4: the facility sets the price and the residue falls on the resident. That asymmetry is the Versorgungslücke.
The Versorgungslücke — the number the product is sold against#
A resident of a Pflegeheim pays, in the vdek’s own presentation, three components R4 R20: the einrichtungseinheitlicher Eigenanteil (EEE) for the care-related cost the SPV contribution does not meet — identical for Pflegegrade 2 to 5 within one facility since 2017 R9, and inclusive of the Ausbildungskosten, which this document previously listed as a fourth head; Unterkunft und Verpflegung; and Investitionskosten, which the vdek notes are “für alle Pflegeheimbewohnenden in einer Einrichtung - unabhängig von der Aufenthaltsdauer - gleich hoch”. Only the EEE is equalised across grades and only the EEE is reduced by the § 43c Leistungszuschläge — 15 % / 30 % / 50 % / 75 % for up to twelve months, more than twelve, more than twenty-four and more than thirty-six months of benefit R4, confirmed word for word in § 43c and restated in the vdek release. The other two are neither capped nor subsidised.
Line |
Amount |
Basis |
|---|---|---|
Average total resident payment, Pflegeheim, first year of stay, 1 January 2026 |
3,245.00 EUR/month, up 261 EUR (9 %) on the year |
|
of which einrichtungseinheitlicher Eigenanteil incl. Ausbildungskosten / Unterkunft und Verpflegung / Investitionskosten |
1,685.00 / 1,046.00 / 514.00 EUR/month |
|
Comparatives: 1 July 2025 / 1 January 2025 |
about 3,108.00 / 2,871.00 EUR/month |
|
Less: net gesetzliche Rente of a median new retiree |
of the order of 1,200.00 to 1,600.00 EUR/month |
|
Residual funded from savings, family or Hilfe zur Pflege |
of the order of 1,650.00 to 2,050.00 EUR/month |
std (A) |
(A) The residual is arithmetic on the two lines above, not an observation; the pension line is still
unsourced. The Eigenanteil figures are no longer the least reliable numbers in the research and
are now among the best sourced — they are read from the vdek’s own evaluation as at 1 January 2026
(gap 15 closed). They are why the market sells Pflegerenten of 1 000 € to 1 500 € a month
unverified, and why delib’s [std] vereinbarte Pflegerente is 1 000,00 € per month. On the
retrieved 2026 level that annuity covers a little under a third of the first-year gap, which is a
smaller share than the figure implied when this document was written against an assumed 3 000 €;
the vereinbarte Rente is a scaling constant and is left where it is, but the reader should not
mistake it for a full solution to the gap.
Two features of the gap matter for the model. It widens over time, because the statutory amounts are uprated episodically — § 30 SGB XI, +4,5 % on 1 January 2025 and nothing again until 1 January 2028 R1 R10 — while the Eigenanteil rose nine per cent in the year to 1 January 2026 alone, the vdek naming “gestiegene Pflegepersonalkosten” as the driver R20. A three-year statutory step against a nine-per-cent annual drift is the case for the Leistungsdynamik option. And it is largest in the first year of a stay, because the § 43c Zuschläge rise with length of stay from 15 % to 75 % R4, so a constant annuity progressively over-covers it; no German wording is known to offer a decreasing care annuity, and delib does not model one.
The three private forms, and why the Pflegerente is one of them#
Pflegetagegeldversicherung |
Pflegekostenversicherung |
Pflegerentenversicherung |
|
|---|---|---|---|
Legal branch |
private Krankenversicherung |
private Krankenversicherung |
Lebensversicherung |
Benefit form |
agreed cash per Pflegegrad, no proof of spend |
reimbursement of a share of residual actual cost |
agreed monthly annuity per Pflegegrad, no proof of spend |
Legal character |
Summenversicherung |
indemnity |
Summenversicherung |
Premium re-rating |
possible under § 203 VVG |
possible |
not possible save on the narrow § 163 VVG route |
Ageing provision |
Alterungsrückstellung where written nach Art der Lebensversicherung; none otherwise |
as Pflegetagegeld |
Deckungsrückstellung always |
Surrender value |
none in substance |
none |
yes, § 169 VVG, subject to the open question below |
Waiver of premium in claim |
usual |
usual |
usual, and contractual |
Death benefit |
rare |
none |
common option |
Pflege-Bahr eligible |
yes — the only eligible form |
no |
no |
Market share / average premium |
dominant / lowest |
negligible / — |
small / highest |
The branch, benefit-form, re-rating and Pflege-Bahr rows follow from R11, R14 and R8 and are structural. The market-share and average-premium row is now sourced. By insured persons at end-2024, Pflegetagegeld is dominant at 3 021 300 against 890 091 subsidised Pflege-Bahr and 366 100 Pflegekosten R21 [S14]; the GDV counts 242 000 stand-alone Pflegerenten in force at end-2023, plus 762 400 written as riders R22. On price, the consumer bodies put the Pflegerente at “etwa zwei- bis dreimal so hoch” as the other forms for comparable cover [S11], and Finanztip at “von Anfang an deutlich teurer” [S12].
The load-bearing difference is the re-rating power. A Pflegetagegeld is health business: MB/EPV § 8b has the insurer compare required against calculated benefits and mortality at least annually per tariff and, on a deviation beyond the “gesetzlich oder tariflich festgelegte[n] Vomhundertsatz”, recalculate every premium in the observation unit [S2] R14, under § 203 VVG R11. A Pflegerente is life business, and the only route is § 163 VVG REG-R27. A buyer at 45 who wants to know what the cover will cost at 80 gets an answer from a Pflegerente and does not get one from a Pflegetagegeld — every consumer comparison reduces to that trade, Finanztip’s included: “ist der Beitrag bei Pflege-Rentenversicherungen für die gesamte Laufzeit festgelegt. Dafür sind diese Tarife jedoch von Anfang an deutlich teurer.” [S12] So does the price difference: the Pflegerente costs more because the insurer carries the basis risk on a fifty-year view of a table built on a superseded assessment regime R15 REG-R51. The second difference is the ageing provision: MB/EPV § 1 Abs. 6 carries an Alterungsrückstellung where the tariff is written nach Art der Lebensversicherung [S2] under § 146 Abs. 1 Nr. 2 VAG R12, and a tariff written without one has a premium that follows attained-age risk upward. The consumer bodies do not put it that starkly: they warn that premiums on every form will rise and still recommend the Pflegetagegeld as the sensible default [S11] [S12], which is the opposite of the reading this document previously attributed to them. The third is the surrender value, the only one of the three from which a policyholder recovers anything on lapse — the Verbraucherzentrale puts it as a positive for the annuity form, that it “kann zudem durch Kündigung beendet werden, ohne dass sämtliche Einzahlungen verloren gehen, wie dies bei einer Tagegeld- oder Pflegekostenversicherung der Fall ist” [S11] REG-R28 — and which makes the contract realisable assets in a Hilfe zur Pflege means test R24.
Market size#
There is a sourced count of German Pflegerente contracts in force, and research gap 12 is closed. This document used to say there was none, on the reasoning that the PKV-Verband counts health-insurance top-up contracts — which it does, and a Pflegerente written by a Lebensversicherer is indeed absent from that series [S16] R21 — and that the GDV life series does not carve the product out. The second half was wrong: the GDV reports Pflegerentenversicherungen as its own line in new business, in-force main covers and riders alike R22 REG-R53.
Measure |
Value |
Year |
Basis |
|---|---|---|---|
Pflegebedürftige in Germany |
5,688,473 (about 5.0 million at end-2021, +730,000 / +15 %) |
end-2023 |
|
Share cared for at home / in vollstationäre Dauerpflege |
85.9 % (4,888,882) / 14.1 % (799,591) |
end-2023 |
|
Projected Pflegebedürftige, constant Pflegequoten |
about 5.6 million (2035) rising to 6.8 million (+37 %) |
2055 |
|
Projected on rising Pflegequoten |
6.3 million (2035), 7.6 million (2055) |
2055 |
|
Private LTC top-up cover, health branch |
3,021,300 Pflegetagegeld; 890,091 Pflege-Bahr; 366,100 Pflegekosten; about 4.5 million persons in all, 5.4 % of the population |
end-2024 |
R21 [S14] |
Pflegerente contracts in force, stand-alone |
242,000 — 0.3 % of 81.4 million life contracts; 177 Mio. EUR annual premium; 29,737 Mio. EUR versicherte Summe |
end-2023 |
|
Pflegerenten-Zusatzversicherungen |
762,400 of 20.7 million riders; 84,218 Mio. EUR versicherte Summe |
end-2023 |
|
Pflegerente new business |
5,499 policies, 0.2 % of life new business by count |
2023 |
Two readings follow from the last three rows. The product is three times more common as a rider than as a stand-alone contract, and this specification models the stand-alone form. And 177 Mio. € across 242 000 contracts is an average in-force premium of about 61 € a month, which is the only external check this document has on the argued premium band below.
Distribution across Pflegegrade R18, read from the Destatis table for end-2023: 13.8 / 40.4 / 29.6 / 11.8 / 4.3 % across grades 1 to 5 — this document previously printed 9 / 44 / 27 / 14 / 6 %, low at grade 1 and high at grades 4 and 5. The reading is unchanged and strengthened: the stock is heavily weighted to the lower grades, which is why a Leistungsstaffel’s middle steps drive its cost. Among residents of Pflegeheime the distribution is quite different — 0.5 / 16.6 / 37.3 / 31.1 / 14.3 %, concentrated in grades 3 to 5 R18 — which is direct support for a grid weighted to the upper grades on a product sold against the residential gap. Age-specific prevalence R18, observed Pflegequoten for 2023: 1.2 % (15–60) · 3.6 (60–65) · 5.5 (65–70) · 9.1 (70–75) · 16.4 (75–80) · 30.8 (80–85) · 53.7 (85–90) · 80.2 (90–95) · 94.5 (95+). This document previously printed a curve about a third to a half lower through the seventies and eighties. Successive five-year ratios are 1.80 · 1.88 · 1.74 · 1.49 · 1.18, so the familiar shorthand — prevalence roughly doubles every five years — is a fair approximation between about 70 and 90 only, and flattens sharply above 90 as the quota saturates. That curve is what the product is built on and the reason a level premium from 45 accumulates for thirty-five years before it starts paying. The Pflegerente is the smallest of the three private forms by contract count and the largest by average premium R22 [S11]; penetration is low because the risk is distant, because Hilfe zur Pflege R24 is a visible backstop, because the Angehörigen-Entlastungsgesetz removed the Elternunterhalt motive for all but high earners from 2020 R24 unverified, and because the products are hard to compare — a point Assekurata makes in its own terms, that buyers overestimate the premium and that “stagnieren die Bestandszuwächse” [S14].
Representative specification#
The representative design is a composite, not a carrier’s tariff. One carrier’s complete wording has now been read — IDEAL Lebensversicherung’s IDEAL PflegeRente Exklusiv, conditions AB-IPR-2022A, with its public Produktbeschreibung beside it [S4] [S5] — so the choices below are no longer argued only against a reconstructed range. One wording is not a market survey, and where the composite differs from it the difference is now marked (differs from [S4]) rather than left unstated. Every choice the corpus still cannot source is a std standardization with a numbered rationale.
Product identity and issue rules#
Parameter |
Representative value |
Basis |
|---|---|---|
Design type |
Individual, single-life, underwritten Pflegerentenversicherung; a stand-alone contract, not a rider on an endowment or a deferred annuity |
[S4] |
Legal branch |
Lebensversicherung, written by a Lebensversicherer; calculated nach Art der Lebensversicherung — level premium, prospective Deckungsrückstellung, no ordinary re-rating |
|
Benefit character |
Summenversicherung: an agreed monthly Pflegerente, paid without proof of expenditure and irrespective of the care setting |
[S4]; setting-independence std (1) |
Trigger |
The statutory Pflegegrad determined under §§ 14, 15 SGB XI, normally by the Medizinischer Dienst for the statutorily insured or MEDICPROOF for the privately insured. The retrieved wording pins the statutory text to a stated version — “Wir beziehen uns immer auf den Stand vom 28.03.2021” — and offers a self-contained Punktesystem as an alternative route |
|
Lives basis and cover period |
Single life — no joint-life Pflegerente is recorded anywhere in the corpus — and whole of life: Versicherungs- und Leistungsdauer lebenslang, the annuity payable for as long as an insured Pflegegrad holds, the contract ending on death |
[S4] [S5] |
Entry ages / age basis |
18 to 65 at entry in the composite, with purchase clustering at 45 to 60; age last birthday at issue, stepping at the policy anniversary. (differs from [S4]: the retrieved tariff writes from 18 to 75, ten years beyond the composite’s envelope) |
[S5]; std (2), (3) |
Vereinbarte Pflegerente |
1 000,00 € per month at Pflegegrad 5; the retrieved tariff’s permitted band is 250 € to 4 000 € a month, against a market band as sold of 1 000 € to 1 500 € |
[S5] for the permitted band; sold band unverified; std (4) |
Currency / sex |
EUR; sex is carried for reporting and for the projection basis, but pricing is unisex for contracts concluded from 21 December 2012 |
|
Pflege-Bahr eligibility |
None, but not for the reason this table used to give. § 127 SGB XI nowhere says Pflegetagegeld. Abs. 2 Nr. 1 requires “die Kalkulation nach Art der Lebensversicherung gemäß § 146 Absatz 1 Nummer 1 und 2 des Versicherungsaufsichtsgesetzes” — the health-insurance provision, whose Nr. 2 requires an Alterungsrückstellung — and Nr. 4 forbids the tariff any benefit beyond the graded Geldleistung. A life-branch Pflegerente forms a Deckungsrückstellung, not an Alterungsrückstellung, and carries surrender and death benefits, so it cannot be a geförderter Tarif |
|
Anchor model cell |
Female, entry age 45, vereinbarte Pflegerente 1 000,00 €/month, |
std (6) |
Footnotes to std rows:
Setting-independence is modern practice. Older wordings paid the full annuity only for vollstationäre care and a reduced one at home unverified; modern practice pays irrespective of setting, which is what makes the product a Summenversicherung. A setting-dependent benefit would need a care-setting state the corpus supplies no transition data for.
Entry ages. The one retrieved tariff writes from 18 to 75 [S5]; the composite keeps 18–65 as its envelope, which is now known to be conservative. It takes 45 as the anchor because that is the lower edge of the observed purchase cluster, not of the permitted band — the two differ by thirty years on the retrieved tariff, and conflating them is the commonest error in describing this product.
No age basis is established for any Pflegerenten tariff; German practice uses the versicherungstechnisches Alter on carrier-specific rounding rules. The composite uses age last birthday at entry, the delib registry’s convention, and a different rule shifts the projection by at most one year of age.
A permitted band is now established [S5]: the retrieved tariff writes a vereinbarte Rente of 250 € to 4 000 € a month, with a minimum of 50 € where only grades 2 and 3 are insured, a minimum annual premium of 60 €, and the ceiling applied to the sum of all Pflegerenten the insurer holds on that life. The band as sold is still unsourced. 1 000,00 € comes from the gap arithmetic above, sits comfortably inside the permitted band, and is a scaling constant — changing it rescales the whole liability linearly. Against the 2026 Eigenanteil of 3 245 € R20 it is a partial cover, not a full one.
Unisex pricing is compulsory from 21 December 2012 REG-R34, and matters more here than on any other delib product: women have materially higher LTC incidence and longer care durations unverified, so the unisex premium embeds a cross-subsidy whose size depends on the sex mix written — itself endogenous to the price. The composite prices on a 50 / 50 blend std and projects on the point’s own sex.
Entry age 45 gives a long enough pre-claim period for the Deckungskapital to be the object it is. The anchor is female deliberately: the projection then runs the higher-incidence basis against a unisex price.
Benefit provisions#
Parameter |
Representative value |
Basis |
|---|---|---|
Benefit |
A monthly Pflegerente, paid in advance, as a percentage of the vereinbarte Pflegerente set by the insured’s current Pflegegrad |
[S4] |
Leistungsstaffel |
0 / 30 / 50 / 75 / 100 % across Pflegegrade 1 to 5 |
std (12) |
Alternative Leistungsstaffel carried |
10 / 20 / 30 / 40 / 100 % — the Pflege-Bahr grid as the market writes it. It is not fixed by statute: § 127 Abs. 2 Nr. 4 SGB XI requires only a Geldleistung at every Pflegegrad, at least 600 € at grade 5, capped at the SGB XI benefit level, with the percentage schedule left to the PKV-Verband’s brancheneinheitliche Vertragsmuster under Abs. 2 Satz 2 — a document this library has not retrieved |
|
Care setting, payment start and duration |
Setting is irrelevant to the benefit — the same annuity is payable at home and in a Pflegeheim; payment starts from the month the assessment fixes as the onset of the insured Pflegebedürftigkeit, subject to any Karenzzeit, and continues to the end of the month in which the grade falls away or the insured dies. The retrieved wording matches on all three points and adds arrears of up to three years |
std (1); [S4] (13) |
Wartezeit (from inception) / Karenzzeit (from onset) |
None / none in the base case, and the retrieved underwritten tariff also has Wartezeit keine [S5]. Observed elsewhere: up to 3 years where present, and § 127 Abs. 2 Nr. 6 SGB XI caps a Pflege-Bahr Wartezeit at “höchstens fünf Jahre”; Karenzzeit commonly none, 3 or 6 months where present |
[S5]; R8 for the five years; other ranges unverified; base std (14) |
Beitragsbefreiung im Leistungsfall |
Full, from the first month in which any annuity is payable; the premium revives on exit from the paying grades |
[S4]; detail std (11) |
Nachprüfung |
The insurer may require periodic evidence that the Pflegegrad persists; the retrieved wording takes that evidence to be the Gutachten of the compulsory-cover carrier, i.e. the statutory determination itself |
[S4] (15) |
Herabstufung / Höherstufung / Reaktivierung |
The annuity moves to the step of the new grade; if the grade falls below the insured threshold, or the insured recovers to the active state, the annuity stops and the premium revives. (differs from [S4]: after 24 months of continuous annuity at grade 4 or 5 the retrieved wording pays for life “auch wenn die Versicherte Person in einen geringeren Pflegegrad eingestuft wird oder die Pflegebedürftigkeit komplett wegfällt” — a lock-in the composite does not carry, and the “guarantee” variant gap 17 asks about) |
[S4] |
Todesfallleistung |
None in the base case; carried as a switchable Beitragsrückgewähr option, which the retrieved tariff writes at 50–80 % of premiums paid on a running premium and 50–100 % on a single premium |
[S4] [S5]; base std (16) |
Leistungsdynamik in payment / Beitragsdynamik before claim |
Off in the base case / not modelled. The retrieved tariff writes a Rentendynamik of 1–5 % a year, and only for the first ten years of the annuity; before the claim, 10 % every three years or 1–5 % a year, “Dynamik endet nach dem 3. Widerspruch in Folge” |
[S5]; std (17) |
Exclusions and territorial scope |
Care caused by war, by the insured’s intentional act and, variably, by addiction, plus any condition excluded at underwriting. The territorial question is now answered for one carrier and the answer is sharper than “worldwide”: cover is worldwide, but a claim arises only if the insured travels “in die EU, in die Schweiz oder nach Norwegen” for the assessment, the same applies to every Nachprüfung, and the contract ends if the insured cannot — a term with real bite for a population that is by definition immobile |
[S4] |
The Leistungsstaffel is the most important parameter in the product and the one the corpus can least support (gap 6). The observed range, attributed to no carrier: grade 1 0–10 %; 2 10–30 %; 3 30–50 %; 4 60–75 %; 5 100 %. A third shape is now known to be written and is not a percentage grid at all: the retrieved tariff sells three product lines that pay the full annuity from grade 2, from grade 3, or from grade 4 — a threshold design [S4] [S5], and the same Stufenmodell structure the DAV’s own bases use, “mindestens Pflegegrad g ist erreicht” R15. Of the two percentage shapes that recur — the Pflege-Bahr 10 / 20 / 30 / 40 / 100 R8, and a flatter, higher shape near 0 / 30 / 50 / 75 / 100, which is what a Pflegerente aimed at the residential gap uses because grades 3 to 5 are where residential care happens — the composite takes the second: grade 1 at 0 % because grade 1 is not a funding event in the statutory scheme either (§§ 36 and 37 both start at grade 2), so insuring it adds incidence-heavy, low-severity claims that dominate the claim count and not the claim cost; grade 2 at 30 %, the top of its range, because that is where care at home begins in earnest; 3 at 50 % and 4 at 75 %, mid-range; 5 at 100 %, the scaling constant. The residential grade distribution — 0.5 / 16.6 / 37.3 / 31.1 / 14.3 % R18 — supports weighting the upper steps on a product sold against the residential gap.
The statutory determination is often backdated, so a wording keyed to the effective date pays earlier than one keyed to the decision date; the composite keys off the effective date, and the retrieved wording does the same — entitlement runs “ab dem Monat, der durch ärztliche Feststellung bzw. den für die Pflegekasse zuständigen Medizinischen Dienst als Zeitpunkt des Eintritts der versicherten Pflegebedürftigkeit festgestellt wird”, with arrears payable up to three years back [S4].
The pairing between underwriting and waiting periods is near deterministic: no underwriting implies a long Wartezeit (the Pflege-Bahr design R8); underwriting implies none. Both are zero in the base run and both are exercised by model points, because a Karenzzeit on a population with heavily elevated mortality removes disproportionately more claims than the same period would on a healthy population.
The insurer does not define the insured event — the state does, and re-defined it in 2017 R9 — and does not assess the claim R6, so claims administration is materially cheaper than on a Berufsunfähigkeitsrente REG-R29; the BGH described the same trade in IV ZR 126/23, the insurer “macht sich den Sachverstand des Medizinischen Dienstes … zunutze und erspart die mit einer erneuten Begutachtung verbundenen Aufwendungen” REG-R36. The price of that cheapness is definition risk — but this document overstated it. A statutory widening does not automatically flow through: every wording retrieved for this product hedges that channel, MB/PPV and MB/EPV by copying the §§ 14–15 test into the conditions [S1] [S2] and AB-IPR-2022A by pinning it to “den Stand vom 28.03.2021” [S4]. What remains unhedged is drift in assessment practice under a fixed text — a loosening of the Begutachtungs-Richtlinien raises incidence with no contractual change and no re-rating remedy — and the reputational and competitive pressure that builds when a pinned private definition parts company with the social insurance beside it.
A Todesfallleistung — most often a Beitragsrückgewähr, written by the retrieved tariff at 50–80 % of premiums paid, or 50–100 % on a single premium [S4] [S5] — converts a pure biometric cover into a savings-bearing contract: it roughly doubles or more the premium for the same annuity std, because the death benefit is close to certain to be paid whereas the annuity is not, and it very likely brings the contract inside the PRIIPs perimeter REG-R32. The base run omits it so that the LTC mechanics are what the model demonstrates.
The Leistungsdynamik is the economically important dynamic, and its cost is counter-intuitively small: the annuity is paid to a population with heavily elevated mortality, so a 2 % escalation on an annuity of about four years’ expected duration costs of the order of 4 % of its value, not the 15 % or 20 % it would cost on a healthy-life pension. The Beitragsdynamik is not modelled at all.
Underwriting and rating#
Parameter |
Representative value |
Basis |
|---|---|---|
Health evidence and question catalogue |
Full Gesundheitsprüfung, with a catalogue materially shorter than a Berufsunfähigkeit application’s: the risk is driven by conditions that predict dependency in old age — cardiovascular and cerebrovascular disease, diabetes, neurological and psychiatric conditions, early cognitive impairment, musculoskeletal disease |
[S4] unverified |
Occupation |
Not a rating factor. The sharpest single contrast with Berufsunfähigkeit |
|
Outcomes |
Accept at standard rates; accept with a Risikozuschlag; accept with a Leistungsausschluss for a named condition; defer; decline |
[S4] unverified |
Absolute bar and disclosure duty |
Existing Pflegebedürftigkeit at application is an absolute bar; the vorvertragliche Anzeigepflicht of § 19 VVG applies, and § 21 Abs. 3 extinguishes the insurer’s remedies “nach Ablauf von fünf Jahren nach Vertragsschluss”, or ten “[h]at der Versicherungsnehmer die Anzeigepflicht vorsätzlich oder arglistig verletzt” — with the bar not running against claims that arose inside the period |
|
Effect of selection on the liability |
Essentially irrelevant to the cost of the benefit. Claims arrive thirty to forty years after underwriting, and the § 19 time bar confines the Gesundheitsprüfung’s effect to the first decade |
|
Sex / smoker status |
Sex may not be a rating factor for contracts concluded from 21 December 2012; smoker status was not established as one for this product |
This is the opposite of Berufsunfähigkeit, where selection is a first-order pricing effect because claims arrive within the working life. Here it matters only to the early-duration reserve, so the model ships no selection factor at all.
Charges#
No charge level of any kind was established for any German Pflegerenten tariff — not one
Abschlusskostensatz, not one administration rate, not one Ratenzahlungszuschlag, not one
Effektivkosten value (gap 2), and no Produktinformationsblatt [S5], Verbraucherinformation
[S7] or Tarifblatt [S9] was located. Every charge in delib is therefore [std]; only the
statutory ceiling is known, and only unverified.
Parameter |
Representative value |
Basis |
|---|---|---|
Abschluss- und Vertriebskosten and the Höchstzillmersatz |
25 ‰ of the Beitragssumme, charged at inception, i.e. exactly at the § 4 DeckRV ceiling: “Der Zillmersatz darf 25 Promille der Summe aller Prämien nicht überschreiten”, cut from 40 ‰ by the LVRG with effect from 1 January 2015. The retrieved wording applies the same rule in the same words — the Verrechnungsverfahren of § 4 DeckRV, “beschränkt [auf] 2,5 % der von Ihnen während der Laufzeit des Vertrags zu zahlenden Beiträge” |
|
Beitragssumme for the Zillmerung base |
Level premium × 12 × (min(premium-end age, 85) − entry age) |
std (20) |
Verwaltungskosten |
3,0 % of each premium collected, plus 2,00 € per policy in force per month at inception prices |
std (19) |
Claims administration / expense inflation |
1,50 € per annuity payment / 1,5 % a year |
std (19) |
Disclosure obligation |
§ 2 Abs. 1 Nr. 1 VVG-InfoV requires the einkalkulierte Abschlusskosten as one total and the übrige and Verwaltungskosten as shares of the annual premium, and Abs. 2 requires them in Euro. The Effektivkosten of Nr. 9 are owed only “bei Lebensversicherungsverträgen, die Versicherungsschutz für ein Risiko bieten, bei dem der Eintritt der Verpflichtung des Versicherers gewiss ist” — the same test § 169 Abs. 1 VVG uses — so a pure-risk Pflegerente carries the euro disclosure and no Effektivkosten figure |
|
Commission |
Not disclosed anywhere in the corpus; folded into the acquisition charge |
std (19) |
No observed range exists for any of these. The levels are round-number placeholders, sized so that the acquisition charge on the anchor is of the same order as the first two years’ premium. Claims administration is set low, the one charge with a real argument behind it: the trigger is determined by a third party R6, so the insurer’s own claims cost is materially smaller than on a Berufsunfähigkeitsrente REG-R29.
The Höchstzillmersatz is a per-mille of the Beitragssumme, the sum of all premiums payable under the contract REG-R16 — not a percentage of the annual premium, and getting that base wrong is a listed pitfall. On a lifelong-premium contract the Beitragssumme is not finite without a convention, so the composite caps the premium term at attained age 85 std, which makes the 25 ‰ ceiling bind visibly rather than notionally.
Termination and values#
Parameter |
Representative value |
Basis |
|---|---|---|
Rückkaufswert |
Payable on surrender. The base measure is the Deckungskapital computed on the premium calculation bases, with acquisition costs spread over at least the first five contract years as a floor on the value, not a cap on the charge |
|
Shipped surrender basis |
A table of guaranteed values expressed as a fraction of premiums paid to date, by completed policy year — the form in which a German contract states them — near zero for the first several years and well below premiums paid for a long time |
std (21) |
Stornoabzug |
0 % in the base run. § 169 Abs. 5 VVG admits a deduction only “wenn er vereinbart, beziffert und angemessen ist”, and makes a deduction for unamortised acquisition costs “unwirksam”. (differs from [S4], and materially: the retrieved wording agrees a Stornoabzug of 25 % of the § 169 value, rising to 50 % after a partial withdrawal, and none on a paid-up conversion) |
|
Beitragsfreistellung |
The policyholder may at any time demand conversion to a paid-up Pflegerente at a reduced vereinbarte Rente, computed from the same § 169 value the surrender path uses |
|
The open statutory question |
§ 169 Abs. 1 VVG owes a Rückkaufswert on “eine Versicherung, die Versicherungsschutz für ein Risiko bietet, bei dem der Eintritt der Verpflichtung des Versicherers gewiss ist” — a positive scope test, not an exception for death-only covers as this document previously described it. On its face a pure-risk Pflegerente does not satisfy it, and § 176 extends §§ 150–170 only to the Berufsunfähigkeitsversicherung. The statutory question is still open; the one wording retrieved answers it in practice by granting a guaranteed § 169 value on a product it calls “eine reine Risikoversicherung ohne Sparprozess” [S4] |
R11; [S4]; research gap 9 |
Lapse and the non-payment path |
Voluntary surrender terminates the contract against the Rückkaufswert less any Stornoabzug; German lapse is in fact a three-way decrement — surrender, Beitragsfreistellung, and premium-default conversion under § 38 VVG, the last two keeping the policy in force with a reduced benefit and a continuing expense loading |
|
Widerruf / expiry |
30 days — § 152 Abs. 1 VVG, “[a]bweichend von § 8 Absatz 1 Satz 1 beträgt die Widerrufsfrist 30 Tage” — with the right lapsing at the latest 24 months and 30 days after conclusion / none: the contract runs for life, with no maturity and no survival benefit |
No Rückkaufswert table for any German Pflegerenten tariff was established. The composite ships guaranteed values as data rather than computing a reserve: § 165 VVG requires the paid-up benefit to be stated in the contract for each insurance year REG-R28, and computing a reserve would break the library’s rule that its models publish gross undiscounted cash flows. The shipped shape encodes the 25 ‰ Zillmerung allowance REG-R16 and the § 169 Abs. 3 five-year spread floor REG-R28.
This footnote was wrong by a factor of five, and it is the sharpest single correction in this document. It said the German life-market Stornoabzug range runs from nil to about 5 % of the Deckungskapital and that none was established for this product. The one Pflegerenten wording since retrieved agrees 25 %, rising to 50 % for a surrender after a partial withdrawal, and justifies it as compensating “die Veränderung der Risikolage des verbleibenden Versichertenbestandes” and the collectively provided risk capital [S4] — a rationale that is specific to a biometric-risk book and has no counterpart on a savings product, which is very likely why the deduction is so much larger here than the life-market shorthand suggests. The base run is still 0 % and the model point still switches it on at 5 %; both are
[std]and neither is now a market figure. Changing them is a model decision and was not taken in this pass.The model implements the surrender path only. Beitragsfreistellung and the § 38 premium-default conversion both keep the policy in force at a reduced vereinbarte Rente, anchored to the same § 169 value REG-R28, and carrying them needs a paid-up ledger for which the corpus supplies no take-up split. The model records the bias: the omitted paths would move policies into a reduced-benefit ledger that still pays claims, so the model understates late-duration claims and overstates surrender outgo.
Contractual mechanics#
The benefit trigger — the statutory Pflegegrad#
The rule. The annuity is payable when, and for as long as, the insured holds a Pflegegrad at or above the lowest grade the Leistungsstaffel pays on, that grade being the one determined under §§ 14, 15 SGB XI by the Medizinischer Dienst for the statutorily insured or by MEDICPROOF for the privately insured R2 R6, with a fallback assessment by a physician the insurer appoints where the insured is covered by neither. The retrieved wording’s fallback is not a physician the insurer appoints but a self-contained Punktesystem offered as an alternative to the SGB XI route, and it pins the statutory route to the §§ 14–15 text “[Stand] 28.03.2021” [S4]. The Verbraucherzentrale records both patterns in the market — insurers “richten sich häufig nach der Einstufung der sozialen Pflegeversicherung. Andere definieren den Leistungsfall nach einer eigenen Systematik” [S11].
What it does. § 14 Abs. 1 defines Pflegebedürftigkeit as “gesundheitlich bedingte Beeinträchtigungen der Selbständigkeit oder der Fähigkeiten” which must persist “auf Dauer, voraussichtlich für mindestens sechs Monate”; § 15 Abs. 2 scores six Module to gewichtete Punkte on a 0-to-100 scale — Mobilität 10 %, cognitive functioning and behaviour sharing 15 % with the higher of the two entering under Abs. 3, Selbstversorgung 40 %, illness- and therapy-related demands 20 %, daily life and social contact 15 % — with thresholds at 12,5 / 27 / 47,5 / 70 / 90 points (§ 15 Abs. 3), all read from the statute. A reader from a US or UK product will recognise an ADL trigger inside the scoring, but the instrument reaches grade 2 on moderate impairments no two-ADL-failure trigger would catch and scores dementia with no physical impairment at all. Three consequences reach the model: the insurer does not define the insured event; the insurer does not assess the claim, which makes claims administration cheap and disputes rare; and a Pflegegrad is a step function of a continuous state, re-assessed episodically R6 — exactly a discrete-state, discrete-time Markov chain. On the first of those, this document used to add that the insurer therefore carries definition risk “no wording can hedge”. That is not right: the wordings hedge the statutory-change channel by writing the definition into the conditions or pinning it to a version date [S1] [S2] [S4], and what is left unhedged is drift in assessment practice under a fixed text.
The 2017 break is still the largest basis risk in the product, but it has been worked. PSG II
(vom 21. Dezember 2015, BGBl. I S. 2424) replaced the three Pflegestufen with the five
Pflegegrade and the time-based assessment with the NBA from 1 January 2017 R9, deliberately
widening the definition — the BGH puts it as “deutlich erweitert”, the § 14 Abs. 2 Nrn. 2 and 3
criteria now scoring what used to count only as erheblich eingeschränkte Alltagskompetenz
REG-R36. The Senat also holds that no inference runs from Pflegegrad 2 back to Pflegestufe
I, the § 140 Überleitung being no help because it moves people into grade 2 with no prior
Pflegestufe at all. What this document did not know is that the actuarial profession answered
the break directly: the DAV publishes a companion Ergebnisbericht re-deriving the DAV 2008 P
Ausscheidewahrscheinlichkeiten for the Pflegegrade, and it prints first-order bases for them
R15. Two things about those bases matter here. They are a Stufenmodell — “mindestens
Pflegegrad g ist erreicht” — not a five-state per-grade chain. And the DAV is candid that “[z]u
Invalidisierungswahrscheinlichkeiten oder gar Invalidensterblichkeiten, für diese fünf Pflegegrade,
fehlt naturgemäß jegliche statistische Information”: the bases are transitions from Pflegestufen
experience, not observations. delib’s transition rates remain an explicitly labelled [std] proxy
shaped on Pflegegrad prevalence, and are not a reproduction of DAV 2008 P.
The Leistungsstaffel#
The rule. The contract fixes one number, the vereinbarte Pflegerente — the monthly annuity at the top Pflegegrad — and a schedule of percentages of it by grade. Everything else in the benefit design is a modifier on that schedule.
What it does. Time spent at each grade is very unequal — a person entering at grade 2 and deteriorating spends most of the spell at grades 2 and 3 and only the final months at grade 5 unverified — so the time-weighted average benefit percentage over a spell is far below 100 %, about 52 % on the profile the research file works through. Two tariffs with the same 100 % top step and different middle steps therefore differ in expected cost by more than the headline suggests, and so does any model that applies an average benefit percentage to an average survival curve. The middle steps carry the cost because the stock is weighted to the lower grades — 13,8 / 40,4 / 29,6 / 11,8 / 4,3 % across grades 1 to 5 at end-2023 R18. What no source supplies is the time spent at each grade: Assekurata says in terms that “keine Informationen darüber [existieren], wie lange die Personen in den einzelnen Pflegegraden verweilen”, the Pflegegrade dating only from 2017, so the 52 % figure above stays an argued profile [S14]. What is sourced is the overall spell: about five years where care begins after 60 — 4,0 for men, 5,7 for women — against about 25 months’ mean stay in a Pflegeheim [S14].
Wartezeit and Karenzzeit#
The rules, and they are two different devices routinely confused in consumer material. A Wartezeit runs from inception of the contract: care beginning inside it is not covered at all, or only where it follows an accident. A Karenzzeit runs from the onset of Pflegebedürftigkeit: the claim is admitted but the annuity does not start until the deferred period has run, and some wordings then pay retroactively to onset and some do not. The retrieved wording has no Wartezeit and pays from the assessed month of onset, with arrears up to three years [S4] [S5].
What they do. The Gesundheitsprüfung does the screening a Wartezeit does in the subsidised product, which is why Pflege-Bahr — § 127 Abs. 2 Nr. 3 SGB XI requires the tariff to waive “eine Risikoprüfung und die Vereinbarung von Risikozuschlägen und Leistungsausschlüssen” — is allowed a Wartezeit of “höchstens fünf Jahre” under Nr. 6 R8, while the underwritten form usually has none, as the retrieved tariff does not [S5]. The mechanic that matters for the model: a deferred period on a population with elevated mortality removes disproportionately more claims than the same period would on a healthy population, because a material share of new claimants die inside it — selection at onset, not at underwriting.
Beitragsbefreiung im Leistungsfall#
The rule. Premiums are waived while the annuity is payable, and the obligation revives if the annuity stops. This is standard for German Pflegerenten and is contractual, not discretionary [S4].
What it does. On a contract issued at 45 and claiming at 82 the waiver removes the remaining premium stream for the whole paying period — of the order of four years of premium, the same order as one year of benefit at the modelled levels. Its cost sits inside the level premium and is one reason a Pflegerente is dearer than a Pflegetagegeld of nominally equal benefit. The interlock with the Leistungsstaffel is the subtle part: waiver runs from the first grade at which an annuity is payable, so on the 0 / 30 / 50 / 75 / 100 grid a life at Pflegegrad 1 is in care, receives nothing, and still pays the premium, while on the Pflege-Bahr grid it is waived — so two tariffs differing only in whether they insure grade 1 differ in premium income as well as benefit outgo, in opposite directions.
Nachprüfung, Herabstufung and Reaktivierung#
The rule. The paying state has three exits, not one. A life receiving the annuity may die; may be downgraded to a lower insured grade, at which the annuity falls to that step; or may be downgraded below the insured threshold or recover to the active state, at which the annuity stops and the premium revives [S4]. Only death is absorbing.
What it does. This is the most important structural fact for an implementation. A model that treats “in claim” as one state exited only by death overstates the liability; one that treats every downgrade as a termination understates it. The model therefore carries the Pflegegrad explicitly and moves lives between grades in both directions. Some tariffs guarantee that an annuity once granted will not be reduced, and the retrieved wording is one of them, on a duration condition rather than outright: twelve months of continuous annuity at grade 4 or 5 make the premium waiver permanent, and twenty-four months make the annuity itself payable for life “auch wenn die Versicherte Person in einen geringeren Pflegegrad eingestuft wird oder die Pflegebedürftigkeit komplett wegfällt” [S4]. Whether that pattern is common is still not established (gap 17), and it matters: on such a wording the paying state is exited only by death after two years, and the state space collapses for the surviving majority of claims. The composite models the unguaranteed form, the more general one; a user holding the guarantee obtains it by setting the recovery and downgrade rates to zero — and, on the retrieved wording’s shape, would want to do so only after the first twenty-four months. The Nachprüfung itself is a documentation exercise rather than the adversarial re-assessment that characterises Berufsunfähigkeit, because the evidence is the statutory determination R6 [S4].
The level guaranteed Beitrag#
The rule. The Beitrag is level and guaranteed for the life of the contract, subject only to § 163 VVG R11 REG-R27. This is the product’s defining commercial property and the whole of its price premium over a Pflegetagegeld.
What it does. The insurer carries the basis risk on a fifty-year view of a table built on a superseded assessment regime, and § 163 is a narrow escape: it needs a non-temporary, unforeseeable change in a calculation basis, an appropriate and necessary new premium and a trustee’s confirmation, and is excluded to the extent the original calculation was insufficient and a diligent actuary should have recognised it REG-R27. The premium must be prudently calculated and permanently sufficient under § 138 VAG REG-R8, which bites hardest on the Pflegewahrscheinlichkeiten. German biometric products are conventionally quoted as a Bruttobeitrag with a lower Zahlbeitrag below it, the gap being a discretionary surplus rebate withdrawable without invoking § 163 at all REG-R27 REG-R53 — so a level Zahlbeitrag is not the same promise as a level Bruttobeitrag. Whether the Pflegerente market quotes the pair as the Berufsunfähigkeit market does was not established (gap 18); the composite models the Bruttobeitrag and no rebate.
The Deckungskapital as an ageing reserve#
The rule. The contract is calculated nach Art der Lebensversicherung, so the reserve is a Deckungsrückstellung under § 341f HGB and the DeckRV R12 R13 REG-R14 REG-R54 — not an Alterungsrückstellung, the private-health-insurance object of § 146 VAG. The precise words are worth using: the Deckungskapital of a Pflegerente is an ageing reserve in economic function and a Deckungsrückstellung in law and in the accounts.
What it does. The annual probability of entering care is negligible before 60, small to 75 and rises steeply thereafter, so the level premium is far above the risk premium for three or four decades and far below it afterwards. Issued at 45, the Deckungskapital rises for roughly thirty-five years, peaks in the early eighties where the incidence curve crosses the level premium, then runs off — later-peaking and smaller relative to premiums paid than an endowment’s, and very much larger than a Risikolebensversicherung’s. Zillmerung up to 25 ‰ of the Beitragssumme REG-R16 produces a negative reserve in the earliest years and a correspondingly poor early-duration surrender value. Two consequences reach the model: interest sensitivity is the highest in delib, because benefits fall on average some thirty-five years after issue; and lapse is profitable early and expensive late, so lapse feeds the premium through the equivalence principle in a real tariff — while the composite’s pricing basis is deliberately lapse-free, because that is German first-order practice and because the house style forbids a pricing quantity that depends on a behavioural assumption that depends on the path that depends on the premium.
Rückkaufswert, Beitragsfreistellung and the Stornoabzug#
The rules. § 169 VVG entitles the policyholder to a surrender value computed as the Deckungskapital on the premium calculation bases, with a floor equal to the value that results from spreading acquisition and distribution costs evenly over the first five contract years — a floor on the value, not a cap on the charge — and a Stornoabzug admissible only if agreed, quantified and appropriate, a deduction for unamortised acquisition costs being expressly ineffective and the burden of proof lying on the insurer REG-R28. § 165 gives an independent right to Beitragsfreistellung at any premium due date, the reduced benefit computed on the same § 169 value and stated in the contract for each insurance year REG-R28.
What they do, and the open question, restated. § 169 Abs. 1 is not an exception for death-only covers — it is a positive scope test: the surrender value is owed on a cover “bei dem der Eintritt der Verpflichtung des Versicherers gewiss ist”. A Risikolebensversicherung falls outside it because the obligation is not certain to arise, and on the face of the words a pure-risk Pflegerente does not satisfy it either; § 176 extends §§ 150–170 entsprechend to the Berufsunfähigkeitsversicherung and to nothing else. The same test governs the Effektivkosten duty in § 2 Abs. 1 Nr. 9 VVG-InfoV, so the disclosure question and the surrender question are one question. The statutory point is still open (gap 9). The one wording retrieved settles it for that carrier the way this document assumed it would be settled: a guaranteed Rückkaufswert “nach § 169 des Versicherungsvertragsgesetzes” with the five-year spread floor, on a product the same conditions call “eine reine Risikoversicherung ohne Sparprozess” [S4]. The composite models a Rückkaufswert, floors it at zero, and exposes the Stornoabzug as a parameter — a parameter the retrieved wording sets at 25 %, not the nil-to-5 % this document assumed.
Überschussbeteiligung#
The rule. The contract participates in surplus under § 153 VVG and § 139 VAG like any other German life contract, unless participation is excluded by agreement R11 R12 REG-R24 REG-R9.
What it does. The composition is different from an endowment’s: there the surplus is dominated
by the Zinsergebnis, while here the reserve is smaller relative to the risk and the biometric
basis is the prudent one, so the Risikoergebnis dominates. That is the Sicherheitszuschlag
between the first- and second-order bases being released as experience emerges REG-R47,
distributed through the Rückstellung für Beitragsrückerstattung under the MindZV and the RfBV
REG-R10 REG-R18 REG-R19. Application forms unverified: Beitragsverrechnung, dominant on
biometric-risk products; verzinsliche Ansammlung; and a Bonus form raising the vereinbarte
Rente. The base run carries no Überschussbeteiligung at all, deliberately: delib publishes
gross undiscounted cash flows, the surplus chassis is demonstrated by
products/kapitallebensversicherung/, and a discretionary Beitragsverrechnung would need a
declared-rate assumption this corpus cannot supply.
Riders and options#
In scope, modelled or parameterized. Leistungsdynamik im Leistungsbezug — escalation of the annuity in payment; the retrieved tariff writes 1 % to 5 % a year and confines it to the first ten years of the annuity [S5], where this document previously assumed 1 % to 3 % and no time limit. Off in the base run, implemented as an escalation ledger running from the first month the annuity is payable; it is the economically important dynamic on this product. Beitragsrückgewähr — a Todesfallleistung returning the premiums paid, off in the base run, implemented in its gross form, without an annuity offset. Abgekürzte Beitragszahlungsdauer, premiums to a fixed attained age, typically 65 or 85 [S5], which the retrieved tariff writes as abgekürzt with a five-year minimum. Einmalbeitrag, offered by the retrieved tariff and combinable with a running premium [S5], explicitly not the base model here but carried so the chassis can price it. Wartezeit and Karenzzeit, both zero in the base run; Risikozuschlag, a multiplier on the gross premium, 1.00 at standard rates; Stornoabzug, 0 % in the base run. And an alternative Leistungsstaffel, the Pflege-Bahr grid as the market writes it — not, as this document previously said, a statutory one R8 — shipped beside the composite’s own so that the effect of the middle steps is demonstrable rather than asserted.
Out of scope, and why. Beitragsdynamik — indexation of premium and cover before claim; the retrieved tariff offers 10 % every three years or 1–5 % a year, ending “nach dem 3. Widerspruch in Folge” [S5] — is not modelled at all: the acceptance rate on each offer is a behavioural assumption this corpus cannot support, and a declined-out contract is an absorbing state needing its own ledger. Überschussbeteiligung is deliberately omitted. Beitragsfreistellung and the § 38 VVG premium-default conversion keep the contract in force with a reduced benefit; the composite treats every voluntary exit as a surrender and records the bias. A Pflegetagegeld or Pflegekosten rider is a different legal branch under a different supervisory regime [S2] R14; Pflege-Bahr is statutorily unavailable R8; bundled assistance packages carry no material cash flow; and no joint-life Pflegerente is in the corpus.
Variations across insurers#
One carrier’s Pflegerenten document has now been read — IDEAL’s IDEAL PflegeRente Exklusiv [S4] [S5] — where none had been before (gap 14). One is not eight, which is still the largest difference between this specification and its frlib counterpart, where eight carriers’ contracts were read and the variation table has eight columns. What follows is therefore still the parameter range the German market is understood to write, with an attribution column that now distinguishes the one column that is attributed from the ranges that are not. Naming a carrier against a value no source supplied would be exactly the fabrication house rule 3 forbids, and no value below is so named except the retrieved one.
Parameter |
Observed / argued range |
Retrieved wording [S4] [S5] |
Basis |
|---|---|---|---|
Leistungsstaffel, grades 1 to 5 |
0–10 % / 10–30 % / 30–50 % / 60–75 % / 100 %; the Pflege-Bahr grid is 10 / 20 / 30 / 40 / 100 %, a market convention and not statutory |
not a percentage grid at all — three product lines paying 100 % from grade 2, from grade 3, or from grade 4; the same Stufenmodell the DAV bases use R15 |
ranges unverified; R8 for what § 127 actually fixes |
Wartezeit, underwritten / Pflege-Bahr / Karenzzeit |
0 to 3 years, usually waived for accident / up to 5 years / 0, 3 or 6 months |
Wartezeit keine |
ranges unverified; R8 for the statutory five-year maximum |
Entry age / purchase cluster / vereinbarte Rente as sold |
18 to 65, some to 70 / 45 to 60 / 1 000 € to 1 500 € per month |
entry 18 to 75; Rente permitted 250 € to 4 000 €, minimum premium 60 €/yr |
ranges unverified; retrieved column [S5] |
Beitragsdynamik / Leistungsdynamik in payment / Stornoabzug |
3 % to 5 % / 1 % to 3 % a year / 0 % to about 5 % of the Deckungskapital |
10 % every 3 yrs or 1–5 %/yr, ending after three refusals / 1–5 %/yr, first ten years only / 25 %, rising to 50 % after a withdrawal |
ranges unverified — and the Stornoabzug range is contradicted by the retrieved wording; retrieved column [S4] [S5] |
Todesfallleistung / Beitragszahlungsdauer |
none, Beitragsrückgewähr, fixed sum or Deckungskapital / lifelong, to 65, to 85 or Einmalbeitrag |
optional Beitragsrückgewähr at 50–80 % of premiums (50–100 % on a single premium) / lifelong or abgekürzt, minimum five years, Einmalbeitrag or a combination |
ranges unverified; retrieved column [S5] |
Benefit by care setting / Herabstufung guarantee / territorial scope |
setting-independent (modern) or reduced at home (older wordings) / present in some wordings, absent in others / not previously established |
setting-independent / guaranteed for life after 24 months at grade 4–5, waiver permanent after 12 / worldwide cover, but assessment and every Nachprüfung must take place in the EU, Switzerland or Norway, failing which the contract ends |
ranges unverified; gap 17; retrieved column [S4] |
What can be said about carriers. The delib brief names twenty-six German undertakings, and this corpus establishes something about exactly one Pflegerenten writer, IDEAL Lebensversicherung a.G. — a Versicherungsverein auf Gegenseitigkeit of Berlin, whose pack names its own conditions AB-IPR-2022A and its supervisory Sicherungsfonds at Protektor Lebensversicherungs-AG [S4]. The structural statement stands: Pflegetagegeld is written by the Krankenversicherer in that list and Pflegerente by the Lebensversicherer, and Stiftung Warentest’s comparison of 70 Pflegetagegeld tariffs from 24 private health insurers contains no Pflegerente at all [S10]. Which other undertakings currently write a Pflegerente is still not established. Where the ranges would have come from: Franke und Bornberg and Morgen & Morgen rate wordings clause by clause — neither was retrieved; Assekurata’s April 2026 study was, and supplies market counts, durations and Pflegetagegeld premiums rather than Pflegerenten clause ranges [S14] REG-R53; Stiftung Warentest and Finanztip publish comparative work, both concentrating on Pflegetagegeld, with the scores paywalled [S10] [S12]; Verivox and Check24 quote on demand and were deliberately not used [S13].
What does not vary. Three things above are legal facts rather than commercial ones and can be stated without attribution: Pflegegrad 5 pays 100 %, which is the definition of the vereinbarte Rente rather than a term; the premium cannot be re-rated outside § 163 VVG, because that is what writing the cover as Lebensversicherung means R11 REG-R27; and a Pflegerente cannot be a geförderter Tarif under § 127 SGB XI, whatever its terms, because Abs. 2 Nr. 1 requires the Alterungsrückstellung of § 146 Abs. 1 Nr. 2 VAG and Nr. 4 forbids any further benefit R8 R12.
Regulatory context#
Social law — the trigger and the first layer. SGB XI creates the soziale Pflegeversicherung
R1; the contribution rate is 3,6 % of assessable earnings from 1 January 2025 and is still
[unverified] here — the rate lives in § 55 SGB XI and was not among the sections read. The
benefit amounts were checked and did not change on 1 January 2026: § 30 Abs. 1 uprated them by
4,5 % on 1 January 2025 and schedules the next rise for 1 January 2028 R1. §§ 14 and
15 define Pflegebedürftigkeit and the five Pflegegrade R2; §§ 36–38 the home-care benefits
R3; §§ 43 and 43c residential care and the Leistungszuschläge R4; §§ 39, 42 and 45b the
secondary heads R5; § 18 and the Begutachtungs-Richtlinien the assessment R6; § 23 the
compulsory private equivalent R7; § 127 the Pflege-Bahr subsidy R8. The reform acts that
matter are PSG II, which introduced the five grades and the einrichtungseinheitlicher
Eigenanteil from 1 January 2017 — vom 21. Dezember 2015, BGBl. I S. 2424 R9 — and the PUEG
of 2023, whose uprating mechanism is now § 30 SGB XI R10. §§ 61–66 SGB XII provide the
means-tested backstop, § 61a carrying the SGB XI definition of Pflegebedürftigkeit across so that
the third layer and the backstop share one trigger R24.
Contract law — the VVG. § 7 and the VVG-InfoV impose the pre-contractual information duties, including the euro disclosure of acquisition and administration costs R11 REG-R31. § 19 governs the vorvertragliche Anzeigepflicht, with § 21 Abs. 3 extinguishing the insurer’s remedies after five years, ten where the breach was intentional or fraudulent R11 REG-R30 — on a product whose claims arrive forty years after underwriting, the time bar is what confines the Gesundheitsprüfung’s effect on incidence to the first decade. § 152 Abs. 1 gives a 30-day Widerruf R11 REG-R23; § 153 the Überschussbeteiligung REG-R24; § 155 the annual Standmitteilung REG-R25 [S8], owed only on contracts carrying an Überschussbeteiligung and listing five items — the benefit on a claim plus surplus, the benefit plus guaranteed surplus on continued and on paid-up terms, the surrender payout, and for contracts from 1 July 2018 the sum of premiums paid — which here reports the guaranteed vereinbarte Pflegerente rather than a sum insured; § 163 the whole of a life insurer’s re-rating power REG-R27; and §§ 165–170 the paid-up conversion, the surrender value, the five-year floor and the Stornoabzug REG-R28. And § 203, the Beitragsanpassung provision that dominates the Pflegetagegeld comparison, applies to health insurance and not to life insurance R11 R14 — which is the point of the entire comparison in this document.
Whether Kapitel 6 VVG reaches this product was not established. §§ 172–177 are the Berufsunfähigkeitsversicherung chapter, and § 177 Abs. 1 extends §§ 173–176 to contracts promising a benefit for a lasting impairment of working capacity REG-R29. A Pflegerente promises a benefit for dependency, not for impairment of working capacity, so on the face of it the extension does not reach it — but the point is unverified, and it matters: § 174’s rule that a cessation of liability takes effect only after notice in Textform and only from the end of the third month following it would, if it applied, put a three-month tail on every Herabstufung and every Reaktivierung. The model does not implement such a tail, and this is the reason.
Supervisory law, reserving and the technical rate. § 138 VAG requires premiums to be prudently calculated and permanently sufficient REG-R8; § 139 governs the Überschussbeteiligung and the Sicherungsbedarf test REG-R9; §§ 140 and 145 the Rückstellung für Beitragsrückerstattung REG-R10, with the MindZV REG-R18 and the RfBV REG-R19 below them; §§ 141–143 create the Verantwortlicher Aktuar and the Treuhänder whose confirmation § 163 VVG requires REG-R11; and § 146 defines the substitutive Krankenversicherung regime, cited only to locate the boundary this product sits on the other side of R14. Above it sits Solvency II, reaching German life business through the VAG REG-R1 REG-R2 REG-R5 REG-R6, with Directive (EU) 2025/2 effective 30 January 2027 REG-R3; nothing here implements a 2027 basis. § 341f HGB requires the Deckungsrückstellung to be computed prospectively on the tariff bases R12 REG-R54. § 2 DeckRV fixes the Höchstrechnungszins, raised to 1,00 % from 1 January 2025 by the Sechste Verordnung of 19 July 2024 — the first increase in about thirty years — after 0,25 % for 2022–2024 and 0,90 % for 2017–2021 REG-R14 REG-R15; the rate applies at contract conclusion and then stays with the contract for its whole term, which is why the German in-force book is a stack of cohorts and why an in-force model point carries its own cohort’s rate rather than today’s. § 4 DeckRV caps the Zillmersatz at 25 ‰ of the Beitragssumme, cut from 40 ‰ from 1 January 2015 by the LVRG REG-R16 REG-R20, and interacts with the independent § 169 VVG five-year floor: the DeckRV governs what the insurer may reserve, § 169 what it must pay, both applying separately with the tighter binding REG-R28. § 5 Abs. 3 DeckRV builds the Referenzzins behind the Zinszusatzreserve REG-R17, which this model does not compute.
Biometric bases. DAV 2008 P is the market-standard first-order basis for German LTC business
on the life chassis R15 REG-R51. It is a multi-state table, supplying
Pflegewahrscheinlichkeiten by sex, attained age and grade of entry, transitions between grades,
Reaktivierungswahrscheinlichkeiten, and, decisively, separate mortality for active lives and for
lives in care, by grade. This document used to say the table is not public. It is: the DAV
publishes the derivation as a free Ergebnisbericht, with the bases themselves in Anhänge 1 to 3,
and a companion report re-deriving them for the Pflegegrade R15. delib still does not
redistribute any of it and no value from it appears anywhere in the library — that is this
library’s own choice, not a licensing constraint. It was built on the pre-2017 Pflegestufen, and
the BGH has held that no inference runs from a Pflegegrad back to a Pflegestufe REG-R36
REG-R51; the profession’s answer was to re-derive rather than to map, on a Stufenmodell
structure and from Pflegestufen data, because for the five grades “fehlt naturgemäß jegliche
statistische Information”. Applying any of it to a five-state per-grade chain is still the
insurer’s own work — the largest single basis risk in the product (gap 10). The DAV’s published
prudence loadings for this risk — an incidence Gesamtzuschlag of 20,5 % to 31,2 % by minimum
grade, an Invalidensterblichkeit Gesamtabschlag of 24,2 % to 28,5 %, and 13,6 % on
Aktivensterblichkeit — are recorded in sources.md and are not implemented here.
Two neighbouring tables enter narrowly: DAV 2008 T for a death benefit written into the contract R16 REG-R48; and DAV 2004 R as a contrast — an annuity table is built to be prudent about people living longer, whereas the annuity here is paid to a heavily impaired population, so using an annuity table would be prudent in exactly the wrong direction and would materially overprice the benefit R16 REG-R49. The German two-basis structure — erster Ordnung for pricing and reserving, zweiter Ordnung for best estimate, the Sicherheitszuschlag the wedge — is at REG-R47, and for care the direction of prudence is higher incidence, longer duration in care and lower mortality of care recipients.
Unisex, tax and disclosure. Sex-based differences in premiums and benefits are prohibited for contracts concluded from 21 December 2012, following Test-Achats and §§ 19, 20 and 33 AGG REG-R34; the tension is sharper here than anywhere else in delib, because the underlying bases are sex-specific and the sex differential in LTC incidence and duration is large. Premiums are sonstige Vorsorgeaufwendungen under § 10 Abs. 1 Nr. 3a EStG, deductible only within an annual ceiling of the order of 1 900 € or 2 800 € — § 10 Abs. 4 Sätze 1 and 2 EStG, the lower figure applying to taxpayers with an employer or public contribution to their health cover R23 — a ceiling the compulsory contributions of Nr. 3 normally exhaust on their own. § 10 Abs. 4 Satz 4 makes the consequence explicit — where the Nr. 3 contributions exceed the ceiling, they are deducted and “ein Abzug von Vorsorgeaufwendungen im Sinne des Absatzes 1 Nummer 3a scheidet aus” — so in practice, for most buyers, the premium is not deductible at all, which is why the Pflege-Bahr Zulage was designed as a direct subsidy R8. The taxation of the benefit is unresolved (gap 13): either exemption under § 3 Nr. 1a EStG, the analysis universally applied to Pflegetagegeld, or Ertragsanteil taxation under § 22 Nr. 1 EStG as a Leibrente, the analysis applied to a Berufsunfähigkeitsrente and the one this product’s life-assurance form argues for R23 REG-R41. § 3 Nr. 1a exempts “Leistungen aus einer Krankenversicherung, aus einer Pflegeversicherung und aus der gesetzlichen Unfallversicherung” without saying whether a life-branch Pflegerente is a Pflegeversicherung for that purpose; the tag stays because the statute is silent on the point and no administrative guidance or authority was retrieved. delib does not model taxation of the benefit and states the open question instead; a Todesfallleistung follows the ordinary life treatment REG-R45 REG-R46. Whether the product is a PRIIP likewise depends on its own design and was not established (gap 16): the Regulation excludes contracts whose benefits are payable only on death or in respect of incapacity REG-R32 unverified, so a pure-risk form falls inside the exclusion while a Beitragsrückgewähr form very likely does not — which makes the presence or absence of a KID in a carrier’s document library evidence about the tariff’s design. The IDD conduct layer applies in any event REG-R33, as does BaFin’s Wohlverhaltensaufsicht strand REG-R35 — though no BaFin material specific to LTC was located (gap 11). The DAV’s Fachgrundsätze bind its members and the DAV makes the annual Höchstrechnungszins recommendation REG-R56; IFRS 17 applies to IFRS reporters with no German carve-out REG-R55; the statutory accounts run on HGB §§ 341–341o, the RechVersV and the BerVersV REG-R54.