Market Ratios & Spreads
One market measured against another, monthly since 2010: whether the major US indices are in a bull or bear regime, the metal ratios, coal and natural gas, the shape of the yield curve, and what bond prices imply about inflation. A long view takes oil/gold back to 1960 and the 10-year yield back to 1990, and small companies are measured against large ones back to 1926. The last chart puts all of them on one scale. Consumer against producer prices has its own sheet: CPI vs PPI.
1Bull or bear: US indices against their 21-month EMA
The rule: a month that closes above its 21-month exponential moving average is a bull month; one that closes below is a bear month. Each chart shows how far an index closed above or below its own EMA (the zero line). All three share one scale. Point at any month to read all three at once.
Bull: close above EMA
Bear: close below EMA
Month-end close vs EMA, %
2Gold / silver
Ounces of silver one ounce of gold buys, monthly average prices. It rises when investors pay up for gold's safety over silver's industrial demand, and falls when growth and risk appetite return.
3Copper / gold and oil / gold
Two industrial commodities priced in gold. Copper/gold is a growth-against-safety gauge that has tended to move with the 10-year Treasury yield; oil/gold isolates energy shocks from general inflation.
Copper ($/lb) ÷ gold ($/oz) × 1000
Oil ($/bbl) ÷ gold ($/oz) × 100
4Energy: coal and natural gas
Thermal coal priced in gold, and natural gas in Europe against the US. They are separate markets: Europe buys much of its gas as imported LNG, so its price follows global LNG supply, while US gas is priced off domestic production. When gas is short, power generators that can switch burn more coal, which is why coal and European gas tend to move together.
Coal ($/tonne) ÷ gold ($/oz) × 100
Natural gas, $ per million BTU
Europe (TTF)
US (Henry Hub)
5The yield curve
Long-term Treasury yield minus short-term, in percentage points. Normally positive; below zero the curve is inverted, which has preceded most US recessions. 2s10s is the market's usual measure; 10-year minus 3-month is the one Federal Reserve research favours.
10-yr minus 2-yr (2s10s)
10-yr minus 3-month
6Inflation expectations and real yields
The 10-year breakeven is the nominal 10-year yield minus the inflation-protected (TIPS) yield: the average inflation the bond market is pricing for the next decade. Set against what CPI actually did, and the real yield an investor locks in after inflation.
10-yr breakeven inflation
CPI, 12-month (realized)
10-yr real yield (TIPS)
7The long view: oil/gold since 1960, the 10-year since 1990
The same oil/gold ratio as above over the whole World Bank record, and the 10-year Treasury yield over the whole Treasury par-curve record, against a 5% line. Gold was fixed at $35 an ounce until August 1971, so before then the ratio mostly tracks the oil price.
Oil ($/bbl) ÷ gold ($/oz) × 100, since 1960
10-year Treasury yield, monthly average, since 1990
10-year yield
5% line
8Small companies against large, since 1926
The smallest 30% of US stocks by market value against the largest 30%, each weighted by size as a fund would hold them, before fees. The first chart grows $1 in each and divides small by large: rising means small companies are pulling ahead. The second shows the 10-year annual return of small minus large, ending each month.
Growth of $1 in small companies ÷ growth of $1 in large companies
10-year annual return, small minus large, percentage points
Context: the small-company claim, tested on data through August 2026
- The claim: a September 2026 investing primer suggested choosing a 401(k)’s small/mid-company fund in place of its large-company fund, on the grounds that over long periods small stocks can do very well and return enough to reach your goals.
- Over the whole record it holds. From July 1926 the smallest 30% returned 11.8% a year against 10.3% for the largest 30%, and $1 in small companies ended with about 3.8 times what $1 in large companies did.
- All of that lead was built by mid-1983. The small÷large ratio peaked at 7.7 in July 1983. It won back much of that ground between 2000 and about 2011, then slid again, and it now stands about half the 1983 peak. Measured from January 1979, small companies trailed: 12.1% a year against 12.5%.
- Small companies won in about two-thirds of all 10-year stretches, but in none ending after March 2019. Over the last 10 years they returned 12.0% a year against 15.8%; over the last 5, 6.9% against 12.4%.
- They were also the bumpier ride: annual volatility of about 21–22% against 15–16% over every window from 5 to 50 years. Research such as Asness, Frazzini, Israel, Moskowitz and Pedersen (2018), “Size matters, if you control your junk”, finds the long-run small-company premium concentrated in profitable, higher-quality small firms rather than in small companies as a group.
- So “enough to reach your goals” is a planning question, not a market one. Small companies compounded about 12% a year over the last decade, which is plenty for many plans. But for 20 years they have been the lower-returning and more volatile choice, so swapping into them meant taking more risk with no return premium to show for it.
Hand-written context; the chart, rows and readout above update from data. These are research portfolios, not a fund: a real small-company fund tracks an index such as the Russell 2000 or S&P SmallCap 600, holds a different slice of the market and charges fees. Reference only, not investment advice.
9Everything on one scale
Ratios, spreads and percentages cannot share one raw axis, so each series is shown as standard deviations from its own 2010–present average (a z-score). 0 is normal for that series; ±2 is unusual. Pick the series to compare.
Method and sources
- Treasury yields: U.S. Department of the Treasury daily par yield curve rates, nominal and real, averaged by month; the tiles show the latest day. Breakeven = nominal 10-year minus real 10-year.
- Gold, silver, copper and crude oil (average of Brent, Dubai and WTI): World Bank Commodity Price Data (the Pink Sheet), monthly averages, CC BY 4.0.
- Coal and natural gas, from the same Pink Sheet: coal is Australian thermal coal f.o.b. Newcastle (6,000 kcal/kg; the futures price since February 2022); European gas is the Dutch TTF hub since April 2015 and an average import border price before that; US gas is the Henry Hub spot price.
- The long view: oil/gold from the Pink Sheet’s full monthly record, which starts in 1960 (gold was fixed at $35 an ounce until August 1971). The 10-year yield is the Treasury daily par-curve rate averaged by month; Treasury publishes that curve from 1990. The current month’s average is month to date.
- CPI and the PPI−CPI gap: BLS, from the data on the CPI vs PPI sheet.
- Bull/bear: month-end closes of the S&P 500, Nasdaq Composite and Dow Jones Industrial Average via FRED, against a 21-month EMA seeded with a 21-month simple average. Index levels are licensed by their publishers and are not shown; only the distance from the EMA is. FRED carries ten years of S&P 500 and Dow history, so their signals start later. The current month’s reading uses the latest close and can change until the month ends.
- Small against large: the value-weighted “Lo 30” and “Hi 30” portfolios formed on size in the Kenneth R. French Data Library (built from CRSP data; NYSE market-value breakpoints, re-formed each June), monthly from July 1926. The library carries a copyright notice and no data licence, so only derived figures are published here: the ratio of the two growth-of-$1 paths, rolling 10-year return differences, and annual return and volatility over a few trailing windows. The monthly returns themselves are not shown. The library updates a month or two behind the market.
- Z-scores use each series’ mean and standard deviation over every month shown. Co-movement on the overlay is not cause. Reference only, not investment advice.