U.S. Inflation & Monetary Dashboard

Look at the CPI decomposition above. Shelter (the orange/red band) has consistently been the largest contributor to headline CPI, accounting for roughly a third of total inflation weight. Notice how food and energy (the more volatile bands) spike and fall quickly, while shelter moves like an ocean liner: slow to accelerate, slow to decelerate.

In 2021–22, energy drove headline CPI sharply higher. By 2023, energy had rolled over, yet headline CPI stayed elevated. The reason is visible in the chart: shelter inflation was still climbing, absorbing the disinflationary impulse from goods and energy. This is the CPI composition effect: what drives the index depends heavily onwhich component you're watching.

Core CPI (ex-food and energy) strips out the volatile components, leaving a signal dominated by shelter and services. When core remains sticky, shelter is usually the reason.

Start with the Fed balance sheet (WALCL). Notice how it expanded sharply in 2008 and 2020, but the M2 response was very different. In 2008, QE created reserves but M2 growth was modest; banks sat on excess reserves. In 2020, fiscal transfers (stimulus checks, PPP) injected money directly into deposit accounts, amplifying M2. Same tool, very different transmission.

The lag chart cross-correlates M2 year-over-year growth with CPI year-over-year growth at various lags. The peak correlation typically falls around 12–18 months, meaning that M2 growth today carries its strongest statistical association with CPI roughly a year to a year and a half later.

Switch the lag chart to Level or Log view to see the raw series side by side. In log view, the slope of M2 steepened dramatically in 2020, the steepest peacetime money growth in decades. CPI's slope steepened about 12–18 months later, consistent with the lag structure.

Rebase everything to Jan 2000 = 100. By now, CPI sits somewhere around 190, meaning consumer prices have roughly doubled in a quarter century. Meanwhile, the S&P 500 is up several-fold, and home prices (Case-Shiller National) have approximately tripled.

This is the two inflations gap: if you hold assets, your net worth has grown far faster than the cost of living. If you don't hold assets, if your wealth is primarily in wages and savings accounts, you've experienced only the CPI side of the equation. The divergence isn't a conspiracy; it reflects how monetary transmission works through financial markets first and consumer prices second.

Note the timing: equities and home prices pulled back sharply during the 2008 financial crisis and again in 2020, but CPI barely dipped. Asset prices are more volatile in both directions. The compounding gap, however, has widened over each successive monetary easing cycle.

The elasticity panel above ranks asset classes by their estimated β to M2. In the baseline sample, the ordering typically looks like: equities > home prices > gold > CPI. That ordering is itself informative: it suggests financial assets absorb monetary impulses more readily than consumer prices.

But look carefully at the confidence intervals. Many of the β estimates overlap, especially for home prices and gold. The ranking is suggestive, not definitive. The bootstrap intervals tell you how much uncertainty surrounds each point estimate.

Now check the cross-validation panel: it tests the same regression across different sample windows. If the equity β is consistently the highest across all windows, that's more convincing. If it jumps around (high in some periods, low in others), that's a sign the relationship is regime-dependent. The instability is part of the evidence.

The rolling-window chart makes this even more visible: you can watch the elasticity estimates evolve over time. They tend to spike during episodes of rapid monetary expansion and compress during calm periods.

The shelter lag chart overlays Case-Shiller home prices (shifted forward ~18 months) against shelter CPI. The fit is striking: turning points in home prices reliably precede turning points in official shelter inflation. Home prices peaked in mid-2022 in year-over-year terms; shelter CPI peaked about a year later.

This isn't a coincidence; it's measurement mechanics. The BLS measures shelter using a rotating panel of rental units. New leases reflect market conditions quickly, but existing leases only reset at renewal. That builds in a 12–18 month smoothing lag. Private rent indices (Zillow, Apartment List) that capture new-lease asking rents show the same leading relationship.

The home price chart below shows the raw Case-Shiller index. Switch to YoY to see the growth rate. The 2020–22 episode was the sharpest home price acceleration since the mid-2000s housing bubble, driven by record-low mortgage rates, pandemic-era savings, and supply constraints. As mortgage rates rose from ~3% to ~7%, home price growth decelerated sharply. That deceleration is now feeding through into lower shelter CPI with the expected lag.

Look at CAPE first. The long-run median for the S&P 500 CAPE is around 16-17x. At the time of writing, CAPE sits well above 30x, in the top decile of historical readings. Only two prior episodes saw CAPE this high: the late-1990s dot-com bubble and the 2021 post-pandemic boom.

But context matters. CAPE is a backward-looking measure (10-year average of past earnings). In a world where real interest rates are structurally lower than the 20th-century average, a higher CAPE may be “fair value” rather than a bubble. The Campbell-Shiller framework says P/E ≈ 1/(r−g): lower discount rates justify higher multiples. The question is whether current rates will persist.

Now compare with Forward P/E, which uses analyst expectations for next-12-month earnings. If Forward P/E is elevated and earnings estimates are flat or declining, that's multiple expansion without fundamental support, a signal that markets may be pricing in rate cuts or liquidity rather than growth. If Forward P/E is rising alongside earnings upgrades, that's a healthier picture.

Price-to-Book is the weakest signal for today's market: the S&P 500 is dominated by tech companies whose value is in intangible assets (IP, network effects, data) that don't appear on balance sheets. P/B still has some information, but interpret it cautiously for the modern index.

The key insight from robustness analysis isn't a single number; it's the pattern of stability and instability across different tests. Here's what tends to hold up, and what doesn't:

• Robust finding: Equities consistently show a higher M2 elasticity than CPI across most sample windows. The ordering (equities > homes > CPI) is more stable than the magnitudes.
• Robust finding: The shelter-CPI lag of 12–18 months is one of the most stable empirical regularities on this dashboard. It holds across different sample periods because it reflects measurement mechanics, not economic regime.
• Fragile finding: The exact M2–CPI peak lag varies by sample. It's around 12–18 months in most windows but can shift to 6–12 months in high-inflation episodes or stretch beyond 18 months in calmer periods.
• Fragile finding: Gold and bond elasticities are highly regime-dependent. Gold was flat in the 1990s and surged in the 2000s–2010s. These estimates should be taken with the widest confidence bands.

When you scroll back through the chapters, pay attention to which patterns survive different sample windows and which don't. That distinction is more informative than any single point estimate.

This dashboard tells a single story through eight chapters. Consumer prices (CPI) are what most people mean by “inflation,” but they're only one channel of monetary transmission, and often the slowest. Asset prices respond faster, move more, and compound over time to create the divergence visible in Chapter 2.

The elasticity analysis (Chapter 4) quantifies that divergence. The housing channel (Chapter 5) shows where asset prices and consumer prices connect, through shelter costs, with a predictable lag. The valuation chapter (6) asks whether current levels are sustainable. And the robustness analysis (7) is honest about what we know and what we don't.

The goal isn't to give you a single answer. It's to give you the tools to think about monetary transmission for yourself, with real data, transparent methods, and honest uncertainty.