Thalassa Capital · May 21, 2026 · Market Commentary
J.P. Morgan’s Global Markets Strategy team just published a thorough framework for estimating 10-year forward returns across both public and private asset classes. It is the kind of rigorous, methodology-first work that tends to get buried in the noise of short-cycle macro commentary — and that is exactly why it deserves a careful read.
The core idea is simple: rather than forecasting returns from the top down, JPM anchors their projections to the Internal Rate of Return (IRR) already embedded in current asset prices. For bonds it is the yield. For equities it is more complex — they back-solve a dividend discount model. For alternatives, they lean on cap rates, vintage-level IRR data, and spread-based forward estimates. The result is a coherent, cross-asset return table that, frankly, I find more useful than most capital market assumption documents out there.
THE HEADLINE NUMBERS
| Asset class | 10yr projected return | Vol | Expected Sharpe | JPM view |
| Global equities | 6.9% | 14.6% | 0.47 | — |
| Global govt bonds | 3.6% | 3.7% | 0.98 | Underweight |
| HG corporate bonds | 4.6% | 5.5% | 0.83 | Overweight |
| HY corporate bonds | 6.4% | 7.5% | 0.86 | Overweight |
| Private equity | 6.9% | 12.8% | 0.54 | Modest UW |
| Real estate | 8.7% | 11.5% | 0.75 | Overweight |
| Hedge funds | 7.2% | 6.2% | 1.15 | Overweight |
| Digital assets | 13.0% | 73.8% | 0.18 | Neutral |
| Private credit | 7.2% | 6.2% | 1.16 | Modest UW |
| Infrastructure | 6.9% | 10.8% | 0.64 | Overweight |
WHAT WE FIND MOST INTERESTING
A few things stand out to us, and not all of them are comfortable for consensus positioning.
On equities: the S&P 500 Equity Risk Premium sits at 2.5% — below its long-run average of 3.2%, but not at the frothy zero we saw at the peak of the dot-com bubble or the 2% lows of 2007. JPM concludes that US equities are not cheap, but they are also not grotesquely overvalued in a historical context. That is a reasonable place to land. The implied equity IRR comes out to 6.9%, which is not a number that screams either opportunity or danger.
The PE premium story has been eroding for a decade — and JPM makes the case in numbers that few want to hear.
On private equity: this is where the paper gets genuinely provocative. The structural arguments for a PE return premium — illiquidity, leverage, multiple expansion — have all weakened materially. The leverage differential between buyout deals and public small caps has compressed. Entry multiples on EV/EBITDA have converged. The secondary market now provides a meaningful exit valve, eroding the illiquidity premium. And empirically, since the GFC the PE return gap over public equities has largely disappeared. Since 2023 it has actually been negative. JPM’s conclusion: pencil in the same 6.9% as listed equities. We think that is directionally right, and more honest than most allocators are willing to be publicly.
On private credit: the 7.2% forward estimate makes sense given current spreads (~475bp in the US) adjusted for defaults and recovery rates. But here is the nuance that matters — that spread premium has compressed as capital has flooded into the asset class. The JPM optimization actually underweights private credit relative to HY bonds, not because of the return estimate, but because HY offers similar economics with better liquidity and useful portfolio correlations. We find that reasoning sound.
On real estate: the 8.7% projected return is the highest in the traditional alternative space, driven by a 6.5% cap rate plus long-run inflation. The catch is capital appreciation — delinquency rates are still rising, extend-and-pretend is alive and well in commercial real estate, and secondary market discounts to NAV remain elevated. The income yield story is solid. The total return story is more qualified.
On hedge funds: JPM projects 7.2%, built from a vol-weighted equity/bond blend plus a 3% alpha assumption. That alpha assumption is grounded in the post-2016 environment, where macro and policy volatility gave hedge funds something to work with. We think the 3% figure is defensible over a full cycle — but it requires manager selection discipline that most allocators either lack or underestimate.
THE PORTFOLIO OPTIMIZATION OUTPUT
Running these estimates through a Black-Litterman framework at a 10% volatility target, the model recommends overweights in HG corporates, infrastructure, HY bonds, real estate, and hedge funds — funded primarily by a large underweight in government bonds and modest underweights in private debt and global equities.
The government bond underweight is the clearest call in the piece. A 3.6% expected return at the cost of duration exposure looks like a poor trade when mid-vol alternatives are yielding 6.9%–8.7%. We might disagree here, but the ultimate outcome is a function of geopolitics and inflation. The inflationary pressures we are witnessing today are eventually going to usher in deflation and the bond story may radically change. Certainly, relative to our beginning of the year view on bonds and duration, a few elements have changed, and we have been forced back into shorter durations.
The digital assets position is interesting: 13% projected return but a 73.8% volatility gives it the lowest expected Sharpe in the table (0.18). The optimizer says hold benchmark weight, which for most institutional portfolios is a rounding error. That strikes us as honest — the return expectation is real, the volatility is real, and the position sizing is a portfolio construction question as much as a conviction question.
OUR TAKEAWAYS FOR THALASSA
The broad message from this framework is one we have been expressing for some time: the risk-adjusted premium available in traditional fixed income has been structurally eroded, while the alternatives landscape has also matured in ways that make the old rule-of-thumb illiquidity premia less reliable than they once were. The edge in alternatives increasingly lives in manager selection, not asset class beta.
We would also push back on the PE return assumption being set equal to public equities rather than slightly below, given the current pricing environment and the 2023-to-present underperformance trend. The convergence argument is directionally right; the symmetric equality assumption may be slightly generous to the asset class.
The JPM report is worth reading in full, particularly pages 5–7 on the private equity structural argument. It is a rigorous and comprehensive pieces of analysis.
Source: J.P. Morgan Global Markets Strategy, “Gauging long-term returns across public and private asset classes,” May 18, 2026. This commentary is for informational purposes only and does not constitute investment advice.