Methodology

What Market Conditions is

Market Conditions describes where a set of market, corporate, and economic measurements currently stand relative to their own history. It is a monitoring tool, not a forecast - see Classifications describe history, not forecasts, below.

Every measurement is calculated the same way, every time, from the same defined sources.

The measurements, and why each is included

Ten measurements are shown, grouped into six categories: Market Behaviour (S&P 500 vs. its 200-day average), Valuation (CAPE), Corporate Economics (FCF margin, FCF conversion, capital expenditures relative to revenue), Financial Conditions (NFCI), Inflation (Core CPI, year-over-year), and Rates (the Fed policy rate, the 10-year TIPS real yield, and the 10-year/2-year Treasury slope).

FCF yield (free cash flow relative to market value) is the tenth measurement; it is not shown historically - see Limitations below.

Historical comparison windows & classification thresholds

Where a percentile classification is shown, it uses these frozen bands: below the 10th percentile is Very Low, 10th–25th is Low, 25th–75th is Typical, 75th–90th is High, and above the 90th is Very High.

Not every measurement receives a percentile classification. The Fed policy rate and the S&P 500 vs. 200-day average are shown with their actual values and no percentile, since a forced classification would not add meaningful context for either. The 10-year/2-year slope treats zero - the inversion boundary - as the primary, economically meaningful reference; its percentile is shown as supplemental context only.

Point-in-time data discipline

Every measurement uses only information that was actually available as of its snapshot date - never information published afterward, even if that later information would have described the same period more completely.

Corporate aggregation & coverage

FCF margin and capex/revenue are computed as an aggregate of free cash flow (or capital expenditures) across the eligible S&P 500 universe, divided by that same universe’s aggregate revenue - equivalent to weighting each company by its revenue, rather than averaging companies equally regardless of size.

FCF conversion is computed differently: aggregate free cash flow divided by aggregate net income across the eligible universe - so it is weighted by each company’s net income, not its revenue.

Financial and Real Estate companies are excluded from all three measurements, since their capital structures make free-cash-flow-based measures not meaningfully comparable to other sectors.

A measurement is only published if it has real data covering at least 80% of the eligible universe’s revenue for that period. Below that threshold, it is withheld rather than shown on incomplete coverage.

The two-fiscal-year staleness rule

A company only contributes to a given measurement if it has a fully aligned, genuinely reported fiscal period within its own two most recent annual reporting periods. This prevents a company with a real reporting gap from contributing multi-year-old figures as if they reflected current conditions.

A company with no sufficiently recent aligned period is excluded from that specific measurement for that period, rather than having an old figure imputed or approximated.

Sources

Corporate measurements are built from company filings via SEC XBRL data. CAPE uses Robert Shiller’s own historical dataset. NFCI, Core CPI, the Fed policy rate, the 10-year TIPS real yield, and the 10-year/2-year slope all come from the Federal Reserve Economic Data (FRED) system. Market price data (the S&P 500 and, for Global Market Context, several international ETFs) comes from Yahoo Finance’s adjusted closing prices, which reflect stock splits and dividend distributions. Global Market Context’s third graph uses RBC GAM’s own published valuation research - see the Global Market Context section below for the distinction.

Historical Reconstruction vs. Published Snapshot

A snapshot labeled Historical Reconstruction was built after the fact, using only information that was genuinely available by its stated cutoff date - it is not something Persistfolio actually published at that time.

A snapshot labeled Published Snapshot is a real, contemporaneous record, created and published on the date it represents. Once published, a snapshot’s stored record is not altered - corrections or later analysis do not silently rewrite what was originally published.

The official monthly snapshot

Persistfolio publishes one official Market Conditions snapshot each month, on the last real market day of that month, using only information available by that cutoff. The live Market Conditions page may reflect more recent information between official snapshots, but the permanent monthly record - what each historical snapshot page represents - is always the frozen, official month-end observation.

Global Market Context methodology

Global Market Context presents three graphs that are built from two genuinely different methodologies - shown together for context, not because they share a data source or calculation approach.

The first two graphs (rolling 3-year returns, and returns relative to the U.S.) use real, tradable ETFs as proxies for each region: SPY for the U.S., EWC for Canada, EFA for developed international markets, and EEM for emerging markets. These use each ETF’s adjusted closing price - Yahoo Finance’s own price series reflecting stock splits and dividend distributions - as the basis for the return calculation, rather than raw, unadjusted prices.

The third graph (valuation relative to equilibrium) is built entirely differently: it reproduces RBC GAM’s own published equilibrium valuation research for the U.S., Canada, Europe, and emerging markets. This graph specifically uses Europe as its own region - not the EFA/EAFE developed-international proxy used in the first two graphs - because that is the actual region RBC GAM’s own published research uses.

Limitations

Historical FCF yield (free cash flow relative to market value) is not available for past dates, because historical market capitalization data is not currently implemented - this measurement is shown as not applicable for any date before it can be genuinely calculated.

Sector-level breakdowns are not part of Market Conditions in this initial version.

Global Market Context is not shown on historical snapshot pages. It was not part of the stored monthly reconstructed dataset those pages are built from, so it is not retroactively attached to them - it is only shown on the live Market Conditions page.

Classifications describe history, not forecasts

Every classification shown in Market Conditions - Very Low through Very High, inverted or positively sloped, above or below a moving average - describes where a measurement currently sits relative to its own history, without implying that reading is good or bad. None of it is a prediction of what will happen next, and none of it is a recommendation to buy, sell, or hold anything.