Inflation breakeven relative value; literature and claim boundary

Economic mechanism

Breakeven inflation is the spread between comparable nominal Treasury and inflation-indexed Treasury yields. It is inflation compensation, not a clean expectation: expected inflation, inflation-risk premia, and relative TIPS liquidity all enter the observed spread. A relative-value identity can therefore be economically coherent only if it names which component it expects to mean-revert or persist and neutralizes duration, carry, index lag, seasonality, funding, and liquidity. A level spread with an attractive chart is not yet a trade.

The local atlas originally named two universes: 2Y/5Y and 5Y/10Y: and three horizons. Those six cells are a search space, not six free trials. Before returns, this review narrows what the held source can actually support: daily 5Y and 10Y inflation-compensation signals plus vintage CPI. It does not support a 2Y leg, inflation swaps, or executable nominal/TIPS baskets.

Primary literature

• Gürkaynak, Sack, and Wright, The TIPS Yield Curve and Inflation Compensation, documents the fitted real-yield curve and shows that inflation compensation is affected by time-varying inflation-risk and early-sample liquidity premia. It supplies measurement foundations, not a trading rule: Federal Reserve research page
• D'Amico, Kim, and Wei, Tips from TIPS, estimates an explicit TIPS-liquidity factor. Their central warning is load-bearing here: treating breakevens as expected inflation without liquidity and risk-premium controls distorts the signal materially: Federal Reserve research page
• Kim, Walsh, and Wei's 2019 update writes the decomposition directly as expected inflation plus inflation-risk premium minus TIPS-liquidity premium and shows the components move through time: Federal Reserve research page
• The Federal Reserve's public TIPS-curve page defines the matched-maturity breakeven and warns that the estimates can be delayed, revised, or changed methodologically. It records a specific 2022 historical revision, so a current download is not itself a historical-vintage archive: Federal Reserve research page
• FRED defines T5YIE and T10YIE as spreads built from constant-maturity nominal and inflation-indexed Treasury series. These are daily market-derived estimates, not security-level transaction prices, cashflows, bid/ask quotes, or total returns: fred.stlouisfed.org record and fred.stlouisfed.org record

What follows for ALPHAC

The literature forbids three flattering shortcuts. First, a breakeven-minus-realized-inflation residual cannot be labelled a pure expectation error. Second, current historical estimates cannot be labelled point-in-time merely because each row has an observation date. Third, changes in a constant-maturity estimate cannot stand in for a tradable, duration/carry-neutral basket without security-level pricing and cashflow evidence.

A later return preregistration must choose one identity before opening prices, define whether it trades cash bonds, inflation swaps, futures, or an explicitly acknowledged ETF proxy, and charge every alternate universe, horizon, sign, and implementation to one family-wise trial account. It must also specify liquidity controls, 2008 stress, indexation lag, seasonality, financing, transaction costs, capacity, DSR/PBO, and the fixed-book diversification tests. No source cited here establishes edge, sign, Sharpe, drawdown, capacity, or admission.