Palanor: Consumer Confidence

Overview

Palanor: Consumer Confidence is a composite read of the American consumer — not their feelings alone, but how those feelings line up against what the household balance sheet is actually doing. The reading is scaled 0 to 100, where 100 is the most confident consumer the model has seen in the 24-month baseline window, and 0 is the most anxious. We watch this number not as a forecast but as an observation: when survey sentiment and credit stress disagree, the index says so.

Why these components

Two survey-based sentiment series anchor the index: the University of Michigan Consumer Sentiment Index and the OECD's normalized Consumer Confidence Indicator. We add CNN Business's Fear & Greed reading as a faster, market-priced sentiment proxy. Against that, we set two hard-data series: the credit card delinquency rate (household stress observed, not asked about) and real retail sales (what the consumer is actually doing with the dollar). The three Google Trends terms — layoff, savings, recession — are the anxiety undercurrent. When households are calm, those searches go quiet; when households are bracing, they rise together.

Methodology

Each component is transformed into a z-score against its trailing 24-month distribution (with the exception of CNN Fear & Greed, which is already a bounded 0–100 series and is recentered around 50, and Retail Sales, which uses a 12-month percentage change because the level series is non-stationary). Components flagged as negative are sign-flipped — rising delinquency lowers the composite. The weighted sum is mapped through a logistic function into the 0–100 range. If a component is missing on a given run (Google Trends rate-limited, CNN endpoint failed), its weight is dropped and the remaining weights are renormalized to sum to 1. The components table records exactly what was used in any particular reading.

Interpretation

Readings above 65 indicate a consumer running confidently above the two-year average. Readings below 35 indicate the household is bracing. The most informative regime is divergence: surveys high, delinquency rising — confidence on paper, stress underneath. The index is built to surface that disagreement rather than mask it.

Caveats

Google Trends is noisy week to week and frequently rate-limited; the index is engineered to renormalize around missing components but heavy reliance on the three search terms is unwise. The 24-month baseline is intentionally short — it reflects recent regimes, not deep history. The index says nothing about why the consumer feels the way they do.

Components & Weights

Every component, every weight. The 30-day spark shows how the underlying series has moved; the transform column shows how it converts to a comparable scale before weighting. Direction indicates whether rising values raise or lower the index.

Awaiting data means the most recent compute did not have a fresh value for that component. The composite re-normalizes around what is present, so the index reading remains meaningful. The component returns the moment its source publishes.

Historical

24-month history of this index. The composite is recomputed daily; values are stored with timestamps.

References

• FRED — Federal Reserve Economic Data, St. Louis Fed. https://fred.stlouisfed.org/
• Atlanta Fed — Wage Growth Tracker and Sticky-Price CPI. https://www.atlantafed.org/research
• Cleveland Fed — Model-based inflation expectations. https://www.clevelandfed.org/our-research/indicators-and-data
• OECD — Composite Consumer Confidence Indicator. https://data.oecd.org/leadind/consumer-confidence-index-cci.htm
• CNN Business — Fear & Greed Index. https://www.cnn.com/markets/fear-and-greed
• Google Trends — interest over time. https://trends.google.com/