Post V: The Assumption

The Obligation Post V of VIII  ·  Forensic System Architecture

The Assumption

The master control of the obligation architecture — a technical input set by interested parties that determines how much of any future obligation is visible in the present, and whose optimism is structurally rewarded and whose accuracy is structurally punished

---

---

Randy Gipe · Claude / Anthropic  ·  2026  ·  Trium Publishing House Limited  ·  Forensic System Architecture

The actuarial report. Dense. Technical. Authoritative. The number on the cover — the reported unfunded liability — is the output of a model whose inputs are assumptions set by the institution whose reported liability depends on those inputs. The number is not false. It is the honest product of assumptions that have been optimistically selected. The document is accurate. The picture it produces is not.

Layer I  ·  Source

---

In 2003, the State of New Jersey issued $2.7 billion in pension obligation bonds. The transaction was designed to close a portion of the state's pension funding gap by borrowing at taxable bond rates and investing the proceeds in the pension fund, earning the assumed investment return of 8.75% annually. If the assumption held, the arbitrage would reduce the unfunded liability. If it did not hold, New Jersey would have both the original pension liability and $2.7 billion in additional bond debt.

The assumption did not hold. The pension fund's actual investment returns in the decade following the bond issuance fell significantly short of 8.75%. New Jersey ended up with both liabilities — the pension underfunding and the bond debt — plus the compounded cost of having made the original pension promise on the basis of an assumption that the institution controlling the assumption had set at a level that made the bond transaction appear viable.

The assumption is not a peripheral element of the obligation architecture. It is the master control. Every figure that appears in every financial document this series has examined — the pension's reported unfunded liability, the bond's coverage ratio, the infrastructure asset's depreciated book value, the deferred maintenance backlog's estimated cost — is the output of a model whose inputs are assumptions. And in every case, the assumptions are set by parties whose institutional interests are served by optimism. The pension fund board that assumes 7.5% annual returns reports a smaller liability than the fund that assumes 5%. The bond issuer whose revenue projections assume robust economic growth shows better coverage ratios than the issuer whose projections are conservative. The infrastructure manager whose condition assessments use the longest plausible useful life assumptions shows the smallest deferred maintenance obligation. The assumption is where the obligation architecture is most completely controlled by the parties who benefit from its invisibility.

Layer II  ·  Conduit

---

The assumption operates as a conduit by translating an unknowable future into a reportable present number. No one knows what investment returns will be over the next thirty years. No one knows what the mortality rate of current employees will be in retirement. No one knows what traffic volumes will support a toll road's revenue bonds in 2045. No one knows how fast a water main will corrode given the chemistry of future source water. These are genuinely uncertain quantities. The assumption converts that uncertainty into a specific number that can be entered into a model and produce a reported liability, a coverage ratio, or an asset value. Without the assumption, the obligation cannot be reported. With the assumption, it is reported — at whatever level the assumption produces.

The Assumption Control Panel — Four Instruments, Four Levers, One Pattern

Who Sets It

Their Incentive

Optimistic Version Produces

Who Bears the Gap

Pension — Discount Rate & Investment Return Assumption

Pension fund board — appointed by the government whose reported liability depends on the assumption.

Lower reported liability. Lower required annual contribution. More budget room for other spending. Political credit for fiscal management.

A pension that appears adequately funded on optimistic assumptions but is significantly underfunded on realistic ones. The gap between 7.5% assumed and 5% actual return compounds silently for decades.

Future taxpayers. Future retirees whose benefit security depends on adequate funding. Future elected officials who inherit the shortfall.

Bond — Revenue Projection & Coverage Assumption

Bond counsel and financial advisors retained by the issuer. Underwriters whose fees depend on the deal closing. Rating agencies responding to issuer-provided projections.

Higher coverage ratios. Better credit ratings. Lower borrowing costs. More debt capacity. Larger bond issuance possible.

A bond that is rated investment grade on revenue projections that prove optimistic, producing debt service coverage shortfalls that require additional tax support or debt restructuring. Puerto Rico: a documented case at scale.

Taxpayers who provide additional revenue when projections fail. Future bondholders if restructuring occurs. Service recipients when revenues must be redirected to debt service.

Infrastructure — Useful Life & Condition Assessment Assumption

Public works departments and asset managers within the government that owns the infrastructure. Depreciation schedules set at asset acquisition, rarely updated to reflect actual condition.

Longer assumed useful life = lower annual depreciation charge = better-looking financial statements. Optimistic condition assessments = smaller reported deferred maintenance backlog = less pressure for capital appropriations.

Infrastructure that appears adequately depreciated in the financial statements while physically deteriorating faster than the accounting reflects. The book value and the repair cost diverge silently until failure forces recognition.

Future taxpayers who fund emergency repairs and reconstruction. Future service users who experience service degradation. Future administrators who inherit failed assets.

Deferred Maintenance — Cost Escalation & Compounding Assumption

Budget offices that do not model deferred maintenance compounding. Capital planning processes that treat each year's deferral as a standalone decision rather than a cumulative obligation.

Each year's deferral appears as a one-time budget saving. The compounding cost of sequential deferrals is never aggregated or reported. The obligation grows in the asset while the budget records savings.

A maintenance backlog that appears manageable in any single year's budget but is catastrophic in aggregate because the compounding has never been modeled. The $180K deferral that becomes a $1.8M reconstruction: multiplied across every deferred asset in every jurisdiction.

Future taxpayers who fund emergency reconstruction. Future service users whose services are cut to fund capital emergencies. Future bond markets that finance reconstruction at debt service costs exceeding the original maintenance budget.

The assumption is set by the party whose reported obligation depends on it. This is not corruption. It is structure — a structural incentive so consistent across every instrument and every jurisdiction that its outputs are predictable before the specific numbers are known.

The Obligation  ·  Series Analysis

Layer III  ·  Conversion

---

What the assumption converts is genuine uncertainty about the future into a reported certainty about the present — and that conversion is always made by the party whose institutional interests are served by the optimistic direction of the conversion. This is the assumption's core conversion function, and it operates identically across all four instruments: the pension fund board that sets the discount rate, the bond issuer whose financial advisor builds the revenue model, the public works department that sets the useful life schedule, and the budget office that treats each year's maintenance deferral as a standalone saving are all performing the same conversion. They are taking an unknowable future quantity and reporting it at the level that best serves the institution's present interests.

1.5–2.5%

The typical gap between public pension fund assumed investment returns and actual realized returns over the past two decades

Analysis of public pension fund investment returns by the Pew Charitable Trusts, the Stanford Institute for Economic Policy Research, and the Center for Retirement Research at Boston College consistently documents that the median public pension fund has earned returns approximately 1 to 2.5 percentage points below its assumed rate over rolling twenty-year periods. A fund assuming 7.5% that earns 5.5% over thirty years is not merely slightly off — the compounded shortfall produces a dramatically larger unfunded liability than the one reported on the optimistic assumption. The gap between the assumed return and the realized return is, over long periods, the primary driver of pension underfunding across the American public pension system.

The Interested Party Problem — Who Sets What, and Why Accuracy Is Punished

The actuary's position

Public pension actuaries are retained by the pension fund board. The board approves the actuarial assumptions. An actuary who recommends significantly lower discount rates — producing a larger reported liability and a higher required contribution — creates political and budgetary pressure that the board, which retained the actuary, must absorb. Actuaries who consistently produce larger liability numbers are retained less frequently than those whose numbers are manageable. The market for actuarial services in the public pension space systematically selects for assumptions that produce results the client can live with.

The bond advisor's position

Financial advisors on bond transactions are paid at closing. A transaction that does not close — because conservative revenue projections produce coverage ratios too low for investment-grade rating — generates no fee. The advisor who builds the revenue model has a structural incentive to produce projections that support the desired rating. The advisor's fee depends on the deal closing. The deal closes if the coverage ratio is adequate. The coverage ratio is adequate if the revenue projection is optimistic enough. The structural incentive is visible before a single projection is built.

The rating agency's position

Rating agencies in the municipal market are paid by issuers — the same issuer-pays model that produced systematic overrating of mortgage-backed securities before 2008. A rating agency that consistently rates municipal issuers below their competitors loses market share. The incentive to accommodate issuer projections and assumptions operates at the ratings level as well as the advisory level. GASB and SEC reforms have improved disclosure requirements, but the fundamental issuer-pays structure of the rating market remains intact.

The elected official's position

The elected official who adopts conservative pension assumptions, requires full actuarial contributions, funds infrastructure maintenance adequately, and reports realistic deferred maintenance obligations faces immediate budget constraints, higher taxes or reduced services, and political criticism from constituencies who benefit from current spending. The costs of fiscal honesty are immediate and politically visible. The benefits of fiscal honesty — lower future obligations, avoided crises, intergenerational equity — accrue to future administrations, future taxpayers, and people not yet born. The elected official who will be in office when those benefits materialize is not the elected official who bears the cost of producing them.

Layer IV  ·  Insulation

---

The assumption's insulation is its technical legitimacy. Assumptions must be made. The future is genuinely uncertain. Actuaries are credentialed professionals applying recognized methodologies. Financial advisors are experienced practitioners with track records. Rating agencies employ analysts with deep domain knowledge. The technical apparatus surrounding the assumption is real — and it provides genuine cover for the systematic bias in the direction the apparatus produces.

The insulation is reinforced by the reform history. GASB Statement 68 required net pension liability on balance sheets — a real improvement in transparency. The Governmental Accounting Standards Board has required more detailed disclosure of actuarial assumptions — a real improvement in legibility. The SEC has increased disclosure requirements for municipal bond official statements. These are genuine reforms that have made the obligation architecture more visible than it was. They have not changed the structural incentives that produce the optimistic assumptions in the first place. The disclosure shows a larger reported liability than was visible before the reform. It shows a larger reported liability calculated on the same interested assumptions that produced the smaller number before.

The assumption is the point where the obligation architecture and the grammar of authority most completely converge. The actuarial report is written entirely in nominalized, passive, technically precise language that presents an interested calculation as a neutral technical finding. "The actuarially determined contribution is calculated based on the entry age normal cost method, using an assumed investment return of 7.25%, applied to the projected benefit obligation discounted at the same rate, producing a net pension liability of…" The passive voice, the nominalization, the defined-term shells — all four mechanisms of the grammar of authority are present in the sentence that reports the obligation that the assumption has made smaller than it is.

Post VI returns to Reedy Creek — the case that makes the entire architecture visible in a single event. A governor tried to use political power to dissolve a special district. The district's bond debt — its outstanding obligation, calculated on assumptions set by its own board, serviced by revenues from a single landowner — proved stronger than the political power trying to override it. The obligation defeated the politics. That is the architecture's ultimate demonstration.

FSA Wall — Post V

The New Jersey 2003 pension obligation bond transaction ($2.7 billion) is documented in state financial records, the New Jersey Division of Pensions and Benefits annual reports, and academic analysis of pension obligation bond outcomes. The subsequent underperformance of the pension fund relative to the 8.75% assumed return is documented in fund annual reports and in academic literature on pension obligation bond risks. The investment return gap between assumed and realized returns (1.5–2.5 percentage points over rolling periods) is documented in research by the Pew Charitable Trusts, the Stanford Institute for Economic Policy Research (Joshua Rauh), and the Center for Retirement Research at Boston College; specific figures vary by measurement period and fund sample. The issuer-pays rating agency model and its relationship to the 2008 financial crisis is extensively documented in the Financial Crisis Inquiry Commission report and subsequent academic literature; the application of similar structural incentives to municipal ratings is the series' analytical inference from documented structural features of the market, not a finding from a specific study. The Puerto Rico debt crisis and the role of optimistic revenue projections in bond issuance is documented in the Financial Oversight and Management Board for Puerto Rico's reports and in academic analysis of the territory's fiscal collapse. The GASB Statement 68 requirement for net pension liability on balance sheets and its 2015 implementation are documented accounting standards. The characterization of the actuarial and financial advisory markets as systematically selecting for optimistic assumptions is the series' analytical judgment grounded in documented incentive structures; it is not an allegation of professional misconduct by specific individuals.

The Obligation  ·  Series Navigation

Post IThe Instrument

Post IIThe Bond

Post IIIThe Pension

Post IVThe Deferral

Post VThe Assumption

Post VIThe District

Post VIIThe Cascade

Post VIIIThe Generation