Generational Accounting¶

Project each birth cohort's net public fiscal flows within a declared government or program boundary and discount them to a common date.

Version: v1 · 2026-10-07 · History
Domain-specific #: 13898
Domain group: Social Sciences
Origin domain: Economics & Finance
Subdomain: Public Finance → Economics & Finance
Aliases: Generational accounts

Core Idea¶

Generational accounting assigns projected public payments and entitlements to birth cohorts under a declared fiscal boundary, then discounts each cohort's dated net stream to a common reference date. The boundary may encompass a government's receipts and transfers or a particular public program. Auerbach, Gokhale, and Kotlikoff present whole-government U.S. accounts in present-value net payments to government; van den Noord and Herd use the approach to value public-pension entitlements and contributions across cohorts in seven major economies. The two applications share the cohort, net-flow, and discounting method, while asking different financing questions.^[1]^[2]

An account is a conditional projection, not an observed lifetime bill or a welfare score. Its value changes with projected demographics, policy, earnings or payment paths, and discount rates. A comparison across cohorts requires the same accounting boundary and an explicit sign convention. In the U.S. article, positive net taxes mean taxes less transfers paid to government; the OECD pension analysis reports net entitlements, with benefits and contributions organized in the opposite interpretive direction. One cannot carry the sign or the U.S. whole-government budget residual into the pension account without a new derivation.^[1]^[2]

Structural Signature¶

Signature: birth cohorts + declared public-fiscal boundary and policy assumptions + projected dated net payment or entitlement streams + present-value conversion → cohort accounts; financing closure and burden comparison are specified uses, not universal membership requirements.

Defined birth-cohort carrier. Assign people to birth periods, including future births where the analysis calls for them. Without cohort assignment, an aggregate fiscal projection is not a generational account.^[1]^[2]
Specified fiscal boundary. Identify the government or program whose receipts, transfers, entitlements, assets, and policy baseline are considered. The U.S. article treats broad government finances; the OECD case treats public pensions. Combining their amounts as if the same boundary were chosen changes the question.^[1]^[2]
Dated cohort net-flow stream. Project payments and receipts over the relevant remaining or full lifetime, and declare which direction counts positive. A one-year transfer or a demographic age ratio alone supplies no lifetime net-flow account.^[1]^[2]
Common-date present value. Discount the future stream under a stated rate or rate path so flows at different dates form a cohort value. The OECD study explicitly varies the rate and treats resulting liabilities as assumption-dependent.^[2]
Variant-specific financing closure. The 1991 U.S. whole-government account uses an intertemporal government budget identity to determine a future-generations residual under its assumptions. The OECD pension account instead presents pension rights, contributions, liabilities, and financing alternatives. Residual closure is not a condition for every generational account.^[1]^[2]
Optional comparison and interpretation. Accounts can compare cohorts or policy scenarios only after their boundaries and assumptions are aligned. Such a comparison is a use of the method; an account can exist without an asserted generational imbalance or welfare ranking.^[1]^[2]

What It Is Not¶

A conventional annual deficit is a flow for a fiscal year, not a discounted projected net-payment stream assigned to birth cohorts. An old-age dependency ratio describes population composition, not taxes, transfers, or pension entitlements for a cohort. A projection of total pension spending is also insufficient if it never allocates net benefits and contributions to cohorts and values them at a common date.^[1]^[2]

Generational imbalance names a possible demographic and fiscal condition; generational accounting is a method for constructing cohort fiscal estimates. An account may be used to investigate imbalance, but the existence of the method does not prove that later cohorts are worse off. Nor do fiscal net payments directly measure lifetime utility, the value of public goods, or the distribution of gains and losses within a birth cohort. The cited originals offer conditional fiscal valuations, not a complete welfare-incidence calculation.^[1]^[2]

Scope of Application¶

The method applies where public finance can be represented as projected dated inflows and outflows attributable to birth cohorts. In the U.S. whole-government version, the authors use NIPA government receipts and expenditures, an age pattern of payments, population-aging projections, and an intertemporal budget constraint. They compare a conditional implied future-generation account with the account of a newborn in 1989. These are features of their particular projection, not a template that every pension or subnational study must copy.^[1]

The OECD paper restricts its applied boundary to public pensions in seven major economies. It assigns projected entitlements and contributions to people retired in 1990, the then-current workforce, living children, and unborn cohorts. It examines pension liabilities and financing alternatives using simplifying assumptions about eligibility, earnings, contribution paths, demographic change, retirement policy, and discount rates. The original PDF's narrative and Annex 2 support this role map; its online copy warns that some tables and figures are unavailable, so this entry does not reproduce uninspected table cells.^[2]

Clarity¶

Ask four questions before interpreting a reported account. Whose cohort? A newborn, existing worker, current retiree, or future birth can have a different observation horizon. Which public boundary? Whole government and a pension scheme contain different flows. Which sign? Net taxes paid to government and net entitlements received from a pension scheme must not be read as if a positive value has the same meaning. Which assumptions? Discounting, growth, future policy, and demographic paths can change the values.^[1]^[2]

The common mechanism is narrower than either paper's policy conclusion: cohort-index the projected fiscal stream, net the payments in a declared direction, and express it in present value. The U.S. residual answers a further whole-government financing question. The OECD's pension liabilities and alternative financing paths answer a program-specific question. Neither extra step is necessary to recognize the underlying accounting method.^[1]^[2]

Manages Complexity¶

Public receipts and spending occur at different ages, in different years, under uncertain future rules. Cohort indexing organizes these events by the people to whom the model assigns them, and discounting brings their projected values to one reference date. The resulting account condenses a long dated stream into a comparable fiscal figure within its own declared model. That compression can expose which assumptions drive a result, but it cannot remove them.^[1]^[2]

A practical audit can therefore separate the calculation into cohort membership, the fiscal boundary, annual net-flow assumptions, the discount rule, and any financing closure. If a conclusion changes when the discount rate or policy baseline changes, the account is still an account; the comparison is conditional. If the cohort flow or present-value operation is missing, the method has changed into a different fiscal statistic.^[2]

Abstract Reasoning¶

Imagine a proposal that shifts a pension contribution from workers now to workers later. A cohort account first assigns projected contributions and entitlements by birth cohort, then discounts each cohort's net pension stream. Only after that valuation can the analyst compare modeled financing arrangements. A current cash transfer to retirees does not alone establish the sign of a cohort's lifetime net entitlement: the OECD discussion distinguishes present cash flows from full projected rights and contributions.^[2]

Conversely, suppose a government reports a smaller annual deficit. That observation alone does not tell how its modeled future cohort accounts changed. Auerbach and colleagues argue, at a theoretical level, that deficit labels can be arbitrary relative to the intergenerational stance of policy. This is the authors' accounting argument, not an empirical finding that all reported deficits are unrelated to fiscal incidence. A generational-accounting analysis would need the relevant cohort flows and assumptions before reaching a comparison.^[1]

Knowledge Transfer¶

The reusable operation transfers from an all-government boundary to a pension-only boundary: declare cohorts, select projected public flows, state the net-flow sign, and discount to a common date. What does not automatically transfer is the U.S. government's residual budget identity, the numerical U.S. result, or the OECD pension entitlement sign. Each applied setting needs its own program rules and financing question.^[1]^[2]

The method may guide questions elsewhere in public finance, but the two cited originals substantiate these two boundaries. Applying it to another program would require original evidence about which payments belong to which cohorts and what assumptions govern them. The general present-value operation is broader than public finance; generational accounting adds cohort and fiscal structure to it.

Examples¶

Canonical: 1991 U.S. whole-government accounts¶

Auerbach, Gokhale, and Kotlikoff present projected accounts for current and future U.S. generations. Their publisher abstract says that, in present value, current and future cohorts' net payments to government and existing government net wealth must be sufficient to finance present-value government consumption. It reports calculations based on NIPA receipts and expenditures and projected population aging. Under its stated range of reasonable growth and interest assumptions, no increase in current generations' projected burden, and equal treatment of future generations except for growth, the abstract reports an implied future-generation fiscal burden 17–24% higher than that of a newborn in 1989. That is a 1991 model projection, not an observed later burden or a portable constant. The original article's full body was not directly inspected for this draft, so no unverified equation or body-page claim is reproduced.^[1]

Mapped back: the cohort carrier distinguishes living, newborn, and future-born groups; the fiscal boundary is the authors' U.S. government receipts, expenditures, wealth, and consumption baseline; the net-flow stream is projected taxes less transfers paid to government; present value makes dated payments comparable at the reference date; the financing closure is the whole-government budget identity that leaves a future-generation residual; the comparison is the conditional future-versus-1989-newborn result. Removing the residual calculation would remove that study's closure inference, while removing cohort-indexed discounted flows would remove generational accounting itself.^[1]

Applied: OECD public-pension cohort valuation¶

Van den Noord and Herd analyze public-pension liabilities in seven major economies. Their Annex 2 assigns prospective pension entitlements and contributions to four cohort groups: those retired in 1990, the 1990 workforce, children living in 1990, and people unborn in 1990. For each it presents entitlements, applicable contributions, and resulting net entitlements as a generational account. Their study examines alternative financing approaches and makes explicit that pension liabilities depend on stylized pension rules, demographic and earnings paths, and the discount rate. The applied boundary is the pension scheme rather than all government receipts and spending.^[2]

Mapped back: the cohort carrier is those four 1990-relative groups; the fiscal boundary is public pension rights and contributions; the net-flow stream uses entitlement less contribution in the paper's sign convention; the present-value rule values future pension streams at a common date; the financing analysis examines program liabilities and alternatives without importing a U.S. whole-government residual; the comparison concerns how modeled financing choices distribute pension obligations among cohorts. A positive cash payment today does not by itself determine the lifetime account.^[2]

Structural Tensions¶

The inspected sources establish a method and conditional applications, but no intrinsic pair of opposed objectives that every generational account must manage. Fiscal perimeter and discount assumptions are declared model choices. A pension-only account and a whole-government account answer different questions; their reported values should be compared only after their boundaries, signs and discount bases are aligned. Diagnostic: Do two reported cohort values use the same fiscal perimeter and sign?^[1]^[2]

Present-value reduction condenses a projected lifetime stream. The OECD paper varies financing assumptions, while the U.S. abstract conditions its result on growth and interest assumptions. A compact figure is interpretable only with those assumptions visible. Diagnostic: Which projected flows and discount choices generated the reported value? These are scope and sensitivity checks, not a universal conflict between precision and comparability.^[1]^[2]

Structural–Framed Character¶

The structural part is substantial: birth-cohort assignment, a specified fiscal boundary, a signed dated stream, and discounting form a reproducible estimation procedure. Removing cohort assignment yields aggregate fiscal projection; removing discounting yields an unvalued sequence rather than the source-defined present-value account. Both unlike applications preserve these roles.^[1]^[2]

Its evaluative weight enters when someone calls a difference burden, Fairness, or generational balance. That judgment depends on policy aims and what the fiscal account excludes, such as utility and within-cohort variation. Human fiscal and accounting practice determines which public receipts and transfers enter the boundary, how flows are attributed to cohorts, and which policy baseline and discount assumptions are projected; those choices frame interpretation without making the present-value operation arbitrary. The method arose in public-finance institutions and travels within that practice when program boundaries change; vocabulary about generations alone is not enough to identify it elsewhere. Importing the word into an unrelated intergenerational story would be analogy unless the cohort fiscal stream and valuation actually appear. Its character: a structural method with framed policy interpretation.

Structural Core vs. Domain Accent¶

The portable core is cohort indexing of a dated flow and present-value reduction under declared assumptions. The strongest all-instance bridge to the wider ontology is the internal Discounting (Present Value) operation. The method's domain accent is the meaning of public receipts, transfers, pension rights, budget boundaries, and birth cohorts. These are not incidental examples: they determine what the calculated account measures.^[1]^[2]

The named identity remains domain-specific because the source-backed instances are public-fiscal cohort accounts, not a demonstrated cross-domain class of all cohort discounted flows. A future Prime would need independently evidenced uses beyond fiscal/public-program accounting with the same necessary role structure. The U.S. budget residual is narrower still: it is one variant's closure, while the OECD pension account keeps the method without it. The accounting core is real, but it does not itself establish a universal welfare ranking or a general law of generational transfer.

This entry is part of Discounting (Present Value).

Strict internal component: Discounting (Present Value). Each admitted account converts future cohort net flows to a common-date value. The Prime can operate on many non-cohort streams; generational accounting adds public-fiscal boundaries and cohort attribution. The parent operation is part of the child method, rather than the child being a subtype of every generic discounting use. This is the sole reviewed strict upward edge.^[1]^[2]

Declined alternatives. A future-flow Projection in ordinary language is involved, but the live Projection Prime denotes a different formal map, so that word match is not a strict edge. The neighboring Generational Imbalance entry describes a condition the method may examine, not a parent kind of method. Fiscal Gap solves for an adjustment to meet a fiscal target; a cohort account need not solve for that adjustment. These distinctions preserve the method's identity and avoid turning every topical connection into a parent.

Relationships to Other Abstractions¶

Current abstraction Generational Accounting Domain-specific

Parents (1) — more general patterns this builds on

Generational Accounting is part of Discounting (Present Value) Prime

Present-value conversion is an internal operation in every admitted generational account.

Hierarchy paths (5) — routes to 3 parentless roots

Generational Accounting → Discounting (Present Value) → Commensurability

Show alternative paths (4)

Neighborhood in Abstraction Space¶

Generational Accounting sits in a sparse region of the domain-specific corpus (88^th percentile for distinctiveness): few abstractions share its structure, so a faithful description tends to retrieve it precisely.

Family — Survival Analysis & Demographic Rates (16 abstractions)

Nearest neighbors

Computed from structural-signature embeddings · 2026-10-08

Not to Be Confused With¶

An annual deficit: a one-year aggregate measure can be discussed without any birth-cohort net-flow valuation. The original U.S. authors propose generational accounts as an alternative perspective and make a theoretical argument about accounting labels; that does not prove a universal empirical nonrelationship.^[1]

Generational imbalance: a possible asymmetric demographic-fiscal condition, which may exist without this computation and is not guaranteed by its output. Old-age dependency: an age-structure ratio, which can help project flows but does not assign lifetime fiscal accounts. Welfare incidence: a normative or utility-sensitive conclusion that a signed net public payment alone cannot settle.^[1]^[2]

A universal future-generations residual: the residual follows from the 1991 whole-government budget identity as specified there. The OECD pension case uses cohort net-entitlement accounting and financing analysis, without making that U.S. closure part of the named method.^[1]^[2]

References¶

[1] Alan J. Auerbach, Jagadeesh Gokhale, and Laurence J. Kotlikoff, “Generational Accounts, A Meaningful Alternative to Deficit Accounting”, Tax Policy and the Economy 5 (1991), 55–110, DOI: 10.1086/tpe.5.20061801. The printed original title uses a colon after “Accounts”; the linked title transcribes it as a comma for the reference binder. Publisher abstract, paragraphs 1–2 and metadata directly inspected; full article body not directly inspected for this draft. registry ↩a ↩b ↩c ↩d ↩e ↩f ↩g ↩h ↩i ↩j ↩k ↩l ↩m ↩n ↩o ↩p ↩q ↩r ↩s ↩t ↩u ↩v ↩w ↩x ↩y

[2] Paul van den Noord and Richard Herd, “Pension Liabilities in the Seven Major Economies”, OECD Economics Department Working Papers 142, OCDE/GD(93)185 (1993), DOI: 10.1787/083510523416. Full original OECD PDF; see Introduction PDF p.4/printed p.5, assumptions PDF pp.11–12/printed pp.12–13, financing discussion PDF pp.28–30/printed pp.29–31, and Annex 2 PDF pp.53–55/printed pp.54–56. The online PDF warns that some tables, graphs, and facsimiles are unavailable; this entry relies on inspected narrative and annex text. registry ↩a ↩b ↩c ↩d ↩e ↩f ↩g ↩h ↩i ↩j ↩k ↩l ↩m ↩n ↩o ↩p ↩q ↩r ↩s ↩t ↩u ↩v ↩w ↩x ↩y ↩z