Capital market assumptions 2026: 18 firms compared.
14 firms provide a US equity point estimate in this table. They range from 3.1 to 9.0 percent a year, with a median of 6.3 percent. These are long-term forecasts, not realised returns. Vanguard's range and forecasts on different horizons are kept out of that median.
Cite: Portfolio Lab, Capital Market Assumptions 2026 comparison, updated 6 October 2026. The CSV contains 17 named-house rows; GMO's seven-year real forecasts are discussed separately below.
Dataset use and source rights
The download is free to access. No blanket open licence is granted for this dataset. Original Portfolio Lab material remains copyright Portfolio Lab; third-party forecasts and source data remain subject to their providers' rights and terms. Cite this page and the original sources, retain their dates and qualifications, and check the relevant rights before redistribution or commercial reuse.
Every year, dozens of the world's largest investment firms publish their capital market assumptions: forward-looking estimates of what each asset class will return over the next decade. Pension funds, endowments, and sovereign wealth funds use these numbers to allocate trillions of dollars.
The Horizon Actuarial Survey collects these estimates from 41 firms, but individual forecasts stay buried behind consulting relationships. The survey tells you the average. It doesn't tell you who said what.
We wanted the full picture. So we downloaded PDFs, read Excel files, and extracted data from bar charts across 18 investment firms whose capital market assumptions are publicly available, from J.P. Morgan and BlackRock to Northern Trust, Schwab and Invesco. Each row names its source edition. Editions and forecast conventions differ; the table preserves missing values and ranges.
Here's where they agree, and where they disagree by as much as 5.9 percentage points.
The 18 Providers
We sourced capital market assumptions from 18 firms that publish 10-year (or similar) forward-looking return forecasts. These range from the largest asset managers in the world (J.P. Morgan, BlackRock, Vanguard) to specialist consultants (Callan, Verus, Meketa) to research-driven firms (Research Affiliates, AQR, GMO).
Together, these firms advise on more than $30 trillion in assets. Their methodologies vary across building-block models, CAPE-based valuation, factor premia and Monte Carlo simulation, which is precisely why comparing them is so revealing.
What 18 Firms Expect from US Equities
US large-cap equities are the single most consequential allocation decision for most investors. Here is what each firm expects, sorted from most bearish to most bullish:
What the houses expect from US equities.
10-YR NOMINAL GEOMETRIC · USD · MOST BEARISH FIRSTThe Horizon Actuarial Survey's 41-firm average for US large cap is 6.5% (2024 edition), against the 14-house mean of 6.2% here. Different samples and publication dates prevent treating that difference as a measured change in expectations.
The Full Picture: Four Asset Classes, 14 Firms
US equities are just one piece. Here's how the same firms see international developed equities, emerging markets, and US bonds, sorted by their US equity forecast:
| Provider | US Eq | Intl Dev | EM | US Bonds |
|---|---|---|---|---|
| Research Affiliates | 3.1% | 7.7% | 7.5% | 5.1% |
| Invesco | 4.7% | 5.9% | 8.5% | 4.5% |
| PGIM | 5.2% | 6.7% | 7.5% | 4.7% |
| Morningstar | 5.3% | 7.5% | 9.9% | 4.5% |
| Verus | 5.4% | 6.8% | 6.7% | 4.7% |
| Schwab | 5.9% | 7.0% | 8.1% | 4.8% |
| AQR | 6.3% | 6.9% | 7.4% | 4.9% |
| Meketa | 6.4% | 7.2% | 7.1% | 4.9% |
| Amundi | 6.5% | n/a | 7.2% | 4.6% |
| J.P. Morgan | 6.7% | 7.5% | 7.8% | 4.8% |
| Northern Trust | 6.8% | n/a | 6.9% | 5.0% |
| Callan | 7.3% | 7.6% | 8.2% | 4.7% |
| BNY Mellon | 7.6% | 7.8% | 8.1% | 4.3% |
| BlackRock | 9.0% | 7.8% | 8.8% | 4.9% |
| Spread | 5.9pp | 1.9pp | 3.2pp | 0.8pp |
| Median | 6.3% | 7.3% | 7.7% | 4.8% |
The pattern is striking. Read the table column by column:
US equities: 5.9 percentage point spread. Wide disagreement. Methodology matters enormously.
International developed: 1.9 percentage point spread. Much tighter. Firms are closer to consensus.
Emerging markets: 3.2 percentage point spread, from 6.7% to 9.9%.
US bonds: 0.8 percentage point spread, from 4.3% to 5.1%.
Where the Published Forecasts Align
1. Comparing international and US equity forecasts
Of the 12 providers that publish both US and international developed point forecasts, the forecasts imply different international return premiums. The average international-minus-US gap is 1.1 percentage points.
The gap ranges from -1.2 percentage points (BlackRock: 9.0% US versus 7.8% international) to 4.6 percentage points (Research Affiliates: 3.1% versus 7.7%). J.P. Morgan's gap is 0.8 percentage points per year: 7.5% minus 6.7%.
These are forecast comparisons, not realised performance. Each firm's horizon, asset definition and method still matter; agreement about direction does not remove uncertainty about the outcome.
2. Bonds are investable again
The tightest consensus in the entire dataset is fixed income. US aggregate bonds: 0.8 percentage point spread across 14 providers (4.3% to 5.1%). After a decade of near-zero yields, the post-2022 rate regime has restored bonds to allocation-worthy status, and everyone agrees on the magnitude.
High yield shows similar agreement: most providers cluster between 5.5% and 6.3%. Cash estimates range from 2.8% (Meketa) to 3.7% (Callan, AQR, Verus), a tight band for an asset class that was near zero just four years ago.
3. Inflation expectations are anchored
Despite pandemic-era inflation shocks, the professional consensus is that long-term inflation will sit near the Fed's 2% target. The Horizon survey shows a 41-firm inflation consensus of 2.4% with a range of just 2.0% to 3.0%. Northern Trust's 2026 report forecasts 2.3% US inflation over 10 years. This is the narrowest spread of any variable in the entire dataset.
Where They Disagree, and Why It Matters
US equities: the biggest debate in finance
The 5.9 percentage point spread on US equities compares the point estimates in our table. GMO's January 2026 US large-cap forecast was -6.5% a year in real terms over seven years. That is a different horizon and inflation convention, so it is not added to the ten-year nominal range.
BlackRock also publishes separate scenario sheets. In its August 2026 workbook, ten-year US equity assumptions range from 2.0% to 15.9%. This is a comparison of scenarios, not a confidence interval or a probability-weighted forecast.
Emerging markets: one major outlier
Emerging-market point estimates range from 6.7% to 9.9%, a spread of 3.2 percentage points. The comparison mixes source editions; differences reflect both the forecasts and when they were made.
Why the Experts Disagree: Three Schools of Thought
The disagreements aren't random. They reflect three distinct philosophies about what drives long-term equity returns:
| Approach | Advocates | US Large Cap | Core Belief |
|---|---|---|---|
| Equilibrium / building blocks | JPM, Northern Trust, Callan | 6.7-7.3% | Revenue growth, margins, and buybacks support returns |
| Valuation-driven | Research Affiliates, GMO | 3.1% nominal / -6.5% real | Starting valuations dominate 7-10yr returns |
| Factor / risk premia | AQR, BlackRock | 6.3% / 9.0% | Systematic premia (equity, value, quality) persist |
The building-block view
J.P. Morgan, Northern Trust, Callan, and most consultants decompose equity returns into building blocks: revenue growth, buyback yield, dividend yield, profit margin change, and valuation impact. Northern Trust uses a modified Grinold-Kroner model; J.P. Morgan uses their own equilibrium framework. These approaches tend to produce forecasts in the 6.5-7.5% range because they anchor to economic growth and corporate fundamentals, which have been structurally positive.
Northern Trust's 2026 report breaks their 6.8% US equity forecast into: revenue growth 3.5%, share change 0%, profit margin expansion 1.6%, valuation contraction -0.5%, and dividend yield 2.0%. They explicitly argue that AI-driven productivity gains justify continued US earnings growth despite elevated valuations.
The valuation view
With the Shiller CAPE ratio near 34, roughly double the long-term average, valuation-focused firms see a powerful headwind. Research Affiliates expects partial mean reversion, yielding 3.1%. GMO takes this further, expecting near-full reversion: their -6.5% real return is a seven-year real forecast, so it cannot be ranked directly against the nominal ten-year figures.
Invesco's 4.7% is their lowest US large cap forecast since they started publishing CMAs in 2017, and it reflects significant valuation caution too. They note that “margins are at record highs which we expect to erode” over the next decade. (For a deeper dive into exactly how these two philosophies produce such different numbers, see our J.P. Morgan vs Research Affiliates comparison.)
The factor view
AQR and BlackRock use risk-premia models that decompose returns into systematic factors: equity risk premium, value, quality, momentum. Their forecasts land in the middle: less constructive than the building-block firms, less bearish than the valuation camp. Both account for valuations but don't assume full mean reversion.
It Gets Wider: BlackRock's Three Scenarios
The August 2026 workbook gives separate Starting point, AI productivity boom and Global risk premia rise scenarios. The table reads their ten-year USD columns. It does not infer scenario probabilities.
| Scenario | US Large Cap | US Small Cap | Private Equity |
|---|---|---|---|
| Starting point | 9.0% | 7.0% | 15.9% |
| AI productivity boom | 15.9% | 7.8% | 27.6% |
| Global risk premia rise | 2.0% | 3.5% | 9.4% |
Source: BlackRock August 2026 workbook, data as of 30 June 2026. Portfolio Lab's workstation currently uses its separately labelled February 2026 BlackRock inputs, with supported inputs adjusted to the app's market-price snapshot. Those app figures need not match this published table.
The Quiet Consensus: Fixed Income
While equities steal the headlines, fixed income shows remarkable agreement across14 providers:
| Provider | US Agg | High Yield | Cash |
|---|---|---|---|
| BlackRock | 4.9% | 6.4% | 3.7% |
| BNY Mellon | 4.3% | n/a | n/a |
| Invesco | 4.5% | 5.7% | n/a |
| Morningstar | 4.5% | n/a | n/a |
| J.P. Morgan | 4.8% | 6.1% | 3.1% |
| PGIM | 4.7% | 4.8% | n/a |
| Verus | 4.7% | 5.7% | 3.7% |
| Callan | 4.7% | 5.7% | 3.7% |
| Schwab | 4.8% | n/a | 3.3% |
| AQR | 4.9% | 5.0% | 3.7% |
| Meketa | 4.9% | 6.3% | 2.8% |
| Northern Trust | 5.0% | 5.5% | 3.3% |
| Research Affiliates | 5.1% | 4.7% | 3.5% |
The 0.8 percentage point spread on US aggregate bonds is narrower than any equity asset class by a factor of two to four. Why? Bond returns are largely determined by starting yields, which are observable. There's less room for philosophical disagreement when the dominant input is a number everyone can see on a Bloomberg terminal.
The Horizon survey confirms this pattern. TIPS has the tightest consensus of any asset class across 41 firms: a spread of just 1.8 percentage points.
How the Consensus Has Shifted
The Horizon survey gives us 5 years of trends. The biggest story is the post-2022 regime change in fixed income:
| Asset Class | 2020 | 2021 | 2022 | 2023 | 2024 |
|---|---|---|---|---|---|
| US Large Cap | 6.2% | 5.8% | 5.9% | 6.8% | 6.4% |
| Intl Developed | 6.8% | 6.4% | 6.6% | 7.5% | 7.1% |
| US Core Bonds | 2.6% | 2.1% | 2.6% | 4.7% | 4.9% |
| Cash | 1.6% | 1.2% | 1.5% | 3.4% | 3.7% |
| Private Equity | 9.1% | 8.8% | 9.1% | 9.5% | 9.2% |
Core bonds went from 2.1% expected return in 2021 to 4.9% in 2024, more than doubling. Cash went from 1.2% to 3.7%. Meanwhile, equity expectations have been remarkably stable. The 60/40 portfolio has been rehabilitated not by higher equity forecasts but by the revival of fixed income.
What This Means for Your Portfolio
1. Don't trust any single forecast
If the 14 firms with US equity point estimates disagree by 5.9percentage points on the most important asset class, your portfolio shouldn't depend on any one number being right. Build for a range of outcomes, not a point estimate.
2. Use the disagreement as a stress test
Run your portfolio through multiple assumption sets. If your retirement plan works under J.P. Morgan's 6.7% US equity forecast but breaks under Research Affiliates' 3.1%, you have a concentration risk problem regardless of who's right. Our CMA comparison tool makes this easy.
3. Over-weight the consensus, not the outliers
The bond forecasts are closer together than the equity forecasts in this survey. Agreement is useful context, but shared assumptions and overlapping methods mean it is not independent proof. Test your plan across the forecast range.
4. Geography matters more than ever
11 of the 12 providers with both US and international forecasts place international developed equities above the US. BlackRock's updated Starting point goes the other way. A majority view is useful context, but it is not unanimous and does not guarantee outperformance.
5. The 60/40 portfolio works again, but differently
With bonds yielding 4-5% instead of 1-2%, balanced portfolios have been rehabilitated. The Horizon survey average for a hypothetical 60/40 pension plan is 7.0% over 20 years, almost exactly the typical institutional return target. But with wide equity uncertainty, diversification beyond a simple 60/40 remains essential.
Explore the data yourself.
Our free CMA comparison tool lets you compare expected returns from J.P. Morgan, BlackRock, Research Affiliates, GMO, and AQR across 57 asset classes, BlackRock's three macro scenarios included.
See every asset class, side by sideFree, no signup required
A Note on Methodology and Comparability
Comparing capital market assumptions across providers requires care:
Time horizon: Most providers in our sample use 10-year horizons (J.P. Morgan, Northern Trust, Schwab, Verus, Callan, Meketa, and others). AQR uses 5-10 years. GMO uses 7 years. Invesco's publication is titled “2025” (10-year from late 2024). Shorter horizons tend to produce more extreme forecasts because starting valuations have a larger impact.
Return type: All figures in this article are nominal geometric (compound annual) returns unless otherwise noted. Some providers publish real returns (GMO is kept separate here). Some publish arithmetic returns (Verus and Invesco publish both; we use geometric).
Asset class definitions: “US Large Cap” maps to S&P 500 for most providers but to MSCI USA for Northern Trust and Russell 1000 for others. “International Developed” is typically MSCI EAFE or equivalent. These differences are generally minor but not zero.
Data sources: J.P. Morgan, BlackRock, Research Affiliates, GMO, and AQR data comes from our CMA comparison tool. The Morningstar cross-firm comparison comes from their “Experts Forecast Stock and Bond Returns: 2026 Edition” Excel dataset. Schwab, Invesco, Northern Trust, Verus, and Meketa data was extracted from their publicly available PDF reports. Callan data was read from their interactive web charts. BNY Mellon, PGIM, and Amundi figures come from publicly available summaries of their 2025/2026 outlook publications.
Effective dates: Most data is from late 2025 / early 2026. The Horizon survey is from early 2024. Meketa and Invesco are titled “2025” (based on late 2024 data). Specific numbers shift year over year, but the structural patterns hold: where the agreement and the disagreement lie is remarkably stable.
Summary
Across the 18 firms in this survey, three distinctions matter:
Bonds are back. With a 0.8 percentage point spread and a 4.8% median, fixed income has been rehabilitated after a decade in the wilderness.
The international premium is not unanimous. 11 of 12 providers with both forecasts favour international. The average expected outperformance is 1.1 percentage points a year in the source tables. This is a forecast comparison, not a realised return.
US equities are genuinely uncertain. The5.9 percentage point spread across 14 firms reflects real intellectual disagreement about valuations, AI-driven productivity, and the sustainability of US corporate margins. Portfolio Lab's Bitcoin input is a separate model estimate; its methodology explains its basis.
This isn't a failure of forecasting. It's an honest reflection of genuine uncertainty. The right response is to build a portfolio that works across the range, rather than to pick the forecast you like best.
Explore the data yourself.
Our free CMA comparison tool lets you compare expected returns from J.P. Morgan, BlackRock, Research Affiliates, GMO, and AQR across 57 asset classes, BlackRock's three macro scenarios included.
See every asset class, side by sideFree, no signup required
Capital Market Assumptions by Firm (2026)
The same numbers, grouped by provider, for readers who arrive looking for one firm's view. Source editions and conventions differ; “n/a” means no comparable point estimate is included here. It does not mean the publisher has no forecast.
Research Affiliates capital market assumptions
US equities 3.1%; international developed 7.7%; emerging markets 7.5%; US bonds 5.1%. Source: Asset Allocation Interactive, January 2026.
Vanguard capital market assumptions
US equities 4.2% to 6.2%; international developed n/a; emerging markets n/a; US bonds n/a. Source: Vanguard Capital Markets Model, run of 30 June 2026.
Invesco capital market assumptions
US equities 4.7%; international developed 5.9%; emerging markets 8.5%; US bonds 4.5%. Source: Capital Market Assumptions 2025, USD.
BlackRock capital market assumptions
US equities 9.0%; international developed 7.8%; emerging markets 8.8%; US bonds 4.9%. Source: August 2026 CMA, data as of 30 June 2026; USD, 10-year Starting point.
PGIM capital market assumptions
US equities 5.2%; international developed 6.7%; emerging markets 7.5%; US bonds 4.7%. Source: Strategic Asset Allocation 2026.
Morningstar capital market assumptions
US equities 5.3%; international developed 7.5%; emerging markets 9.9%; US bonds 4.5%. Source: 2026 Forecast.
Verus capital market assumptions
US equities 5.4%; international developed 6.8%; emerging markets 6.7%; US bonds 4.7%. Source: Capital Market Assumptions 2026.
Schwab capital market assumptions
US equities 5.9%; international developed 7.0%; emerging markets 8.1%; US bonds 4.8%. Source: Long-Term Capital Market Expectations 2026.
Amundi capital market assumptions
US equities 6.5%; international developed n/a; emerging markets 7.2%; US bonds 4.6%. Source: Capital Market Assumptions 2026.
AQR capital market assumptions
US equities 6.3%; international developed 6.9%; emerging markets 7.4%; US bonds 4.9%. Source: Capital Market Assumptions 2026.
Meketa capital market assumptions
US equities 6.4%; international developed 7.2%; emerging markets 7.1%; US bonds 4.9%. Source: Capital Market Expectations 2025.
Northern Trust capital market assumptions
US equities 6.8%; international developed n/a; emerging markets 6.9%; US bonds 5.0%. Source: Capital Market Assumptions 2026.
Callan capital market assumptions
US equities 7.3%; international developed 7.6%; emerging markets 8.2%; US bonds 4.7%. Source: Capital Market Assumptions 2026.
BNY Mellon capital market assumptions
US equities 7.6%; international developed 7.8%; emerging markets 8.1%; US bonds 4.3%. Source: Capital Market Assumptions 2026: Endurance Under Pressure.
Fidelity capital market assumptions
US equities n/a; international developed n/a; emerging markets n/a; US bonds n/a. Source: Capital Market Assumptions, 20-year horizon (2025-2044).
PIMCO capital market assumptions
US equities n/a; international developed n/a; emerging markets n/a; US bonds n/a. Source: PIMCO Capital Market Assumptions, semiannual, five-year horizon.
Sources
- J.P. Morgan Long-Term Capital Market Assumptions, 2026 (30th Edition)
- BlackRock Capital Market Assumptions, 2026 (3 scenarios)
- Northern Trust Capital Market Assumptions, 2026 Edition (released Dec 2025)
- Schwab Long-Term Capital Market Expectations, 2026
- Invesco Capital Market Assumptions, 2025 (USD)
- Verus Capital Market Assumptions, 2026
- Meketa Capital Markets Expectations, 2025 (Condensed)
- Callan Capital Market Assumptions, 2026 (interactive chart)
- Research Affiliates Asset Allocation Interactive, January 2026
- GMO 7-Year Asset Class Forecast, January 2026
- AQR Alternative Thinking 2026 Issue 1: Capital Market Assumptions
- Morningstar “Experts Forecast Stock and Bond Returns: 2026 Edition”
- BNY Investments Capital Market Assumptions 2026 (Endurance Under Pressure)
- PGIM Strategic Asset Allocation Assumptions, 2026
- Amundi Capital Market Assumptions, 2026
- Vanguard Capital Markets Model (VCMM), 2026
- Horizon Actuarial Survey of Capital Market Assumptions, 2024 Edition (41 respondents, free download)
Every house, in detail
One page per firm, each stating the edition its figures came from, the method behind them, and where the published document sits. Sorted most cautious first on US equities.
Figures are the 10-year nominal US equity forecast. “n/p” means the firm publishes assumptions but not a point estimate for US equities.
All houses in one table →Glenn Cameron, CFA
Founder, Portfolio Lab. 25+ years in institutional portfolio management.