Methodology · sensitivity analysis
The Reality Index headline uses a fixed 1980 household basket — a Laspeyres index. Applied to the same corrected price data, other defensible weighting schemes produce different and generally larger gaps. This page holds the prices fixed and varies only the weights, to show how much the weighting choice alone moves the result.
Why this page exists. Critics will argue that any inflation index has cherry-picked its methodology. Our answer is to hold the price data fixed and vary only the weights, so readers can see how much the weighting choice alone moves the gap — and that the gap between Reality Index and official CPI persists across the whole range. The headline (18.5% cumulative gap, fixed 1980 basket) is the most conservative of the five. Other defensible weightings push the gap up to about 65% over the same 45-year window.
The principle. We do not switch to whichever scheme produces the largest gap — that would be the dishonest move. We anchor the headline to the fixed 1980 basket precisely because it produces the smallest gap of the defensible schemes: if even the most conservative weighting shows official CPI understating the 1980 cost of living, the finding cannot be dismissed as cherry-picking. The explorer exists to show the range, not to pick the highest number.
Headline gap range, 1980 → 2025
Across five defensible weighting schemes, the Reality Index gap over CPI ranges from +18.5% (the headline fixed 1980 basket) to +64.6% (BEA PCE weights, used by the Federal Reserve). The headline sits at the conservative floor — we anchor to it deliberately. The retired 2024-weights backcast produces +39.5% on the same corrected prices — reproducing the index’s original published headline almost exactly, which confirms the move from the original 39.5% to today’s 18.5% headline was the weighting choice, not the price data.
Choose a weighting basis — the headline recomputes
The headline. Each category is weighted by its 1980 share and held fixed across every year — a Laspeyres index, and the most conservative of the defensible bases.
The same underlying bucket inflation indexes — what families paid for housing, food, healthcare, transportation, and so on — produce a wide range of headline rates depending on how the categories are weighted. This is true of every inflation index, including CPI itself. CPI's headline number is fundamentally a weighting decision applied to thousands of underlying price observations. Change the weights, change the headline.
Reality Index anchors the headline to a fixed 1980 basket. We considered modern BLS CES weights — the December 2024 Relative Importance shares — but applying today’s basket to 1980 prices over-weights categories that only grew large later, which inflates the historical gap. The honest choice is to fix the basket at the start of the window and let prices move. The gap we are flagging is in the price measurement; the weighting choice, if anything, makes our headline more conservative than the modern-weights alternatives.
The alternative weighting scenarios on this page exist to address the legitimate question: what if BLS's CES weights themselves understate certain categories? The National Academies Panel on Redesigning the BLS Consumer Expenditure Surveys has explicitly stated that underreporting in CES "appears to differ sharply across commodities, raising the possibility of differential biases in the Consumer Price Index." This is not a fringe critique. It is a mainstream methodological concern documented by BLS's own academic advisory body.
The BEA PCE scenario is particularly worth understanding. PCE weights are what the Federal Reserve actually uses for monetary policy decisions. PCE weights healthcare differently than CPI does (PCE includes employer-paid premiums in the healthcare bucket, which CPI does not, because CPI is supposed to capture out-of-pocket consumer prices). When you apply the Fed’s own preferred weighting scheme to our retail price data, the cumulative gap rises to about 65%. The Fed’s own weighting methodology, applied to real retail prices, produces an inflation rate roughly 65% higher than the official CPI over the period.
The takeaway
The 18.5% cumulative gap between Reality Index and official CPI is the conservative answer. It uses the most conservative defensible weighting scheme (a fixed 1980 basket) and the most independent price data available (KFF, NCES sticker tuition, FHFA, EIA, AAA, BLS Average Price Data).
Including the headline, the five schemes span roughly +18% to +65%. The gap exists under every one. What changes is the magnitude. What does not change is the direction: under every reasonable weighting, the goods and services families actually buy have inflated faster than the official Consumer Price Index reports.
That is the finding. The methodology choices that follow from it — which gauge the Federal Reserve should use, how COLA adjustments should be computed, what the right reference rate is for contracts and pensions — are policy debates Reality Index does not attempt to resolve. We just publish the range honestly.