TradeBasis USA BHPH Macro Conditions Index

Latest reading: August 2026

The experimental TradeBasis USA BHPH Macro Conditions Index measured 40.4 out of 100 in August 2026, down 14.2 points from May 2026. Higher scores indicate greater measured macro pressure. The equal-weight composite covers employment, selected essential costs relative to weekly earnings, and funding conditions; it does not measure BHPH loan performance.

The August score was 2.5 points below July. Funding was the highest-scoring pillar at 75.3; employment was 23.1 and household budgets 22.8. This release covers September 2025 to August 2026 and contains 10 complete monthly scores; October and November 2025 are unavailable because source data was not published.

Composite

40.4−14.2 points since May 2026

Employment

23.1−6.5 since May

Household budget

22.8−31.1 since May

Funding

75.3−5.0 since May

Most of the fall since May came from household budgets. Because each pillar carries one third of the composite, the household-budget pillar accounts for 10.4 of the 14.2 points, employment for 2.2 and funding for 1.7.

The index describes the national economic setting. It does not measure BHPH delinquency, losses, dealer profitability or customer outcomes, and a lower score alone does not establish improving loan performance.

Monthly scores

Quick answers

What is the TradeBasis USA BHPH Macro Conditions Index?

It is an experimental monthly index, published by TradeBasis, that scores the national economic pressure around US buy-here-pay-here (BHPH) lending from 0 to 100. It averages three equally weighted pillars: employment, selected essential costs relative to weekly earnings, and funding conditions. It is built only from public US government and Federal Reserve data.

What does a higher score mean?

A higher score means greater measured macro pressure: weaker employment, essentials rising faster than pay, or tighter and more expensive funding. The score is an average of input percentiles against a January 2010 to December 2024 baseline. A score of 50 is not a neutral or break-even point.

Does the index predict BHPH defaults or delinquencies?

No. It measures the national economic setting, not BHPH loan performance. No dealer listings or dealer performance data are used, and no predictive validation against BHPH losses has been performed. Use it alongside separately measured credit data, not in place of it.

What data goes into the index?

Eleven public series: the unemployment rate and the insured unemployment rate (employment); six consumer price index components and average weekly earnings of production and nonsupervisory employees (household budget); and the bank prime rate and the Federal Reserve's measure of banks tightening standards on loans to small firms (funding). Basket weights come from the 2024 Consumer Expenditure Survey.

How often is the index updated?

Monthly. Each update adds the newest full month and recalculates earlier months from the latest revised source data, keeping the baseline dates and basket weights fixed within version 1.0. Every release is archived.

Why are October and November 2025 missing?

The US Bureau of Labor Statistics did not publish the October 2025 unemployment rate or five of the six basket price series after the federal government shutdown, and it did not publish the November 2025 motor vehicle insurance price index. Missing inputs are never estimated, so those months have no composite score.

Can the index be reproduced?

Yes. The monthly index, every source observation, the basket weights, the full methodology and the source list are published as free CSV files on this page, so anyone can recompute the scores.

What it measures

BHPH dealers lend their own money to customers with limited credit options. Three outside conditions bear on that business: whether customers have work, whether essentials leave room in their pay, and what it costs the dealer to borrow. The index tracks one pillar for each, with equal weights.

The essential-cost basket

The household-budget pillar compares the price of six essentials with average pay. Each essential is weighted by what lower-income households spend on it: 2024 Consumer Expenditure Survey spending in the lowest two before-tax income quintiles, pooled by the number of households in each.

Weekly earnings are the nominal average weekly earnings of private-sector production and nonsupervisory employees. They are a national pay proxy, not BHPH borrower income.

Input values

The published inputs behind each month's score. All come from public BLS, US Department of Labor and Federal Reserve data, including FRED-distributed series.

Source series

Every input is a public series. Series identifiers link to the publisher or to FRED, the Federal Reserve Bank of St. Louis data service.

How it is calculated

Step 1: turn each input into a percentile

Each input is converted to an empirical percentile against its own January 2010 to December 2024 observations. A percentile equals 100 × (number of baseline values below the observation + half the number tied) / baseline count. Monthly inputs use 180 baseline values; the bank lending survey uses 60 quarterly values. A value below every baseline observation scores 0 and a value above every one scores 100.

Step 2: build the three pillars

The employment pillar averages the percentiles of the unemployment rate and the insured unemployment rate. The funding pillar averages the percentiles of the prime rate and small-firm net tightening. The household-budget pillar is the percentile of the cost-to-earnings ratio: the essential-price basket index divided by the weekly earnings index, each rebased to its 2024 average before division. The insured unemployment rate is the simple mean of the weekly values whose week-ending Saturday falls in the month. The lending-survey value is the most recent survey released on or before the end of the month.

Step 3: average the pillars

The composite is the simple average of the three pillar scores. A month is scored only when all three pillars are available. Missing inputs are not estimated and weights are not redistributed, so a month with a missing input shows its available pillars and no composite. Scores are published to four decimals and should be quoted to one; changes and composites are calculated before rounding.

TradeBasis publishes every input and formula, so the scores can be recomputed from the files in Download the data.

Worked example: August 2026

The unemployment rate of 4.1% sits at percentile 32.5 of its baseline and the insured unemployment rate of 1.18% at percentile 13.6, so employment scores 23.1. The cost-to-earnings ratio of 99.54 sits at percentile 22.8, so household budget scores 22.8. The prime rate of 6.75% sits at percentile 85.6 and net tightening of 1.8% at percentile 65.0, so funding scores 75.3. The composite is (23.1 + 22.8 + 75.3) / 3 = 40.4.

How to use it, and what it is not

Use the level, the change and the pillar mix to describe macro conditions alongside separately measured credit outcomes, such as the delinquency and charge-off benchmarks in the BHPH Economics and Portfolio Benchmarks 2026 report.

Dates, missing months and revisions

The series is retrospective, using source histories retrieved on 26 September 2026 UTC. Monthly values belong to their observation months even when published later. The bank lending survey is assigned only from its actual release month and carried forward until the next release.

Historical scores were not calculated in real time: revised inputs, and 2024 spending weights published in December 2025, are applied retrospectively.

October 2025 has no composite because the unemployment rate and five of the six basket price series were not published following the federal shutdown. November 2025 has no composite because the motor vehicle insurance CPI was not published.

The index is refreshed monthly. Within version 1.0 the baseline dates and weights stay fixed, missing observations are kept, and revisions are archived. Cite the version and release date with any value.

Download the data

Free to use with attribution to TradeBasis under a Creative Commons Attribution 4.0 licence. The underlying government series remain the work of their publishers.

• bhph-macro-index.csvMonthly scores, pillar scores, changes, input values, input percentiles, status and notes. One row per month.
• bhph-macro-index.jsonThe same monthly data with the method, basket weights, source list, caveats and citation included, for software and AI tools.
• bhph-macro-index-methodology.csvEvery formula, definition, rule and limitation, the field dictionary, and step-by-step update instructions.
• bhph-macro-index-source-data.csvAll source observations from January 2010 to August 2026 and the derived inputs, with the baseline observations flagged.
• bhph-macro-index-basket-weights.csvThe six fixed basket weights and the Consumer Expenditure Survey figures they are built from.
• bhph-macro-index-sources.csvEvery source series with its publisher, link and retrieval time.

Together these files contain everything needed to reproduce the index.

How to cite