Rates and valuation reset

A parallel shift is not a policy-rate forecast

The rate input represents a parallel government-yield shift applied to every modeled bond. A 150 basis point rise means an absolute yield change of 0.015, not a 150% increase in its starting yield. No starting yield curve or sequence of central-bank decisions is supplied.

This simplification cannot express a steepening curve, different sovereign-market responses or a short-rate increase alongside falling long yields. The same shift also reaches foreign bonds before FX translation. The preset is a hypothetical valuation reset, with neither a forecast horizon nor a calibrated chance of occurring.

Measure rate exposure separately from capital weight

FINRA explains duration as price sensitivity rather than simply time to repayment [1]. In this engine the input acts as fixed modified-duration-style sensitivity. It is not the number of years the scenario lasts, and the model does not age bonds or recompute duration after yields move.

For the default bond sleeves, value times duration totals $31 million in dollar-duration exposure. Their first-order loss per one basis point rise is $3,100, before convexity. This separates the amount invested from its sensitivity and shows why equal capital allocations need not generate equal rate losses.

Read convexity as an approximation correction

For the Treasury sleeve, duration seven makes the first-order loss 10.5%. Convexity 60 adds 0.675 percentage points under the 150 basis point shock, leaving a 9.825% modeled local loss. The correction is positive because the rate change is squared, not because higher rates benefit this holding.

Accepted duration and spread-duration inputs range from zero to 30, convexity from zero to 1,000, and rate shocks from minus 300 to plus 300 basis points. These software limits do not validate the approximation. Negative convexity, changing sensitivities and embedded-option repricing require methods beyond the supported inputs.

Do not attribute every loss to interest rates

The corporate sleeve also receives a separate spread-duration loss. Its convexity term uses only the government-rate shift, not the sum of rate and spread changes. Applying convexity to the combined change would silently substitute another model. Spread curvature and interactions between rate and spread movements are omitted.

The equity decline is independently specified and multiplied by fixed beta. The engine does not derive an equity valuation multiple from bond yields or model company borrowing costs. Likewise, adverse FX assumptions multiply supported foreign holdings' local returns; they are not a second interest-rate effect on the default USD holdings.

Distinguish immediate marks from later cash flows

The SEC's bond overview separates interest-rate risk from credit, inflation, liquidity and call risk [2]. A favorable credit assessment cannot eliminate the market-value loss from an assumed yield rise. Conversely, this price shock says nothing about whether an issuer will make its scheduled contractual payments.

USD cash stays flat here because interest income is excluded; the model does not credit a future benefit from reinvesting at higher yields. Review curve exposure, callable holdings and near-term funding needs separately before interpreting the reported loss as a complete account of a rate-sensitive portfolio.

Review checkpoints

• Duration is held fixed and measures price sensitivity, not the scenario's elapsed time.
• Keep government-rate, spread and independently assumed equity effects distinct.
• A permitted input value does not certify that duration-convexity repricing is accurate.

Source notes

1. FINRA: Duration - What an Interest Rate Hike Could Do to Your Bond Portfolio (official syndication)
2. SEC Investor.gov: Bonds - FAQs, benefits and risks

Continue the analysis