top of page

Is Bitcoin a Good Diversifier? What JPMorgan and BlackRock Actually Disagree On

Oct 21
6 min read

Updated: Oct 26

The answer depends on one number nobody can pin down, and on a second one you control.

In October 2026, J.P. Morgan Asset Management published a long-term return forecast for Bitcoin and, in the same document, declined to fully include it in the models it uses to build portfolios. Two months earlier, BlackRock had published a paper arguing that a 1–2% allocation would have improved a classic 60/40 portfolio over the past decade.

Read quickly, that looks like a bank saying no and an asset manager saying yes. It's narrower than that. Both firms accept that Bitcoin has a return worth forecasting. They disagree about whether its relationship with everything else in a portfolio is stable enough to model. That relationship is the whole question of Bitcoin diversification, and it's one you can measure in your own holdings rather than take on faith.


Bitcoin Diversification

What each firm said

J.P. Morgan. Its 2027 Long-Term Capital Market Assumptions add a Bitcoin return estimate but keep the asset out of the full covariance matrix, the table that tells a portfolio model how assets move relative to one another. The stated reasons are volatility and unstable correlations. In plain terms: JPMorgan is willing to guess what Bitcoin might return, but not to let an optimizer treat its past co-movement with stocks and bonds as reliable.

BlackRock. Re-Underwriting Bitcoin (August 2026) runs a hypothetical backtest from May 2016 to May 2026. It takes a US 60/40 portfolio and moves 1% or 2% from equities into Bitcoin. Both versions show better risk-adjusted returns than the original. The paper is upfront about the limits: the results are backward-looking, exclude fees, and the period contains episodes when Bitcoin's correlation with risk assets rose sharply.

Put side by side, the two positions aren't contradictory. BlackRock shows what happened over one ten-year window. JPMorgan questions whether the next window will look the same. Both are right about what they're claiming.


Why correlation decides almost everything

Whether a small Bitcoin position diversifies or simply adds risk depends mostly on how it moves with the rest of the portfolio. Here is a simplified example that makes this concrete.

Assume a portfolio of 58% equities, 40% bonds and 2% Bitcoin. Give equities 16% annual volatility, bonds 6%, Bitcoin 60%, and assume stocks and bonds are uncorrelated. Now vary only one thing: Bitcoin's correlation with stocks.

BTC weight

BTC–equity correlation

Portfolio volatility

Share of total risk from BTC

2%

0.0

9.7%

1.5%

2%

0.3

10.0%

4.8%

2%

0.6

10.3%

7.6%

5%

0.0

9.6%

9.8%

5%

0.3

10.4%

15.7%

5%

0.6

11.1%

20.1%

Illustrative assumptions, not forecasts. Risk share is each asset's contribution to portfolio variance.

Two things stand out.

At zero correlation, a 2% Bitcoin position contributes less risk than its weight, and total volatility actually falls slightly compared with the higher-correlation cases. That's diversification working as the textbook describes.

At a correlation of 0.6, the same 2% position is responsible for nearly 8% of the portfolio's risk, and a 5% position for about a fifth of it. Nothing about Bitcoin changed except how closely it tracked stocks.

This is exactly where JPMorgan's objection bites. If the correlation input swings between those rows depending on the year, any allocation built on a single historical estimate is fragile.


Bitcoin's correlation with stocks depends on when you look

Correlation runs from -1 to +1 and describes how two return series move together over a specific sample. Change the sample and you can get a different answer.

Three choices drive the result:

  • The benchmark. Bitcoin against the S&P 500 is not the same as Bitcoin against the Nasdaq 100 or a global equity index.

  • The frequency. Daily returns pick up short-term, liquidity-driven moves. Weekly or monthly returns smooth them out.

  • The window. A 90-day correlation can look nothing like a three-year one.

Rolling correlations, where you recompute the figure every day over a moving window, show how much the relationship drifts. A single long-run average hides that. The more useful question for most investors is narrower: what did Bitcoin do in the weeks when equities were already falling? A low average correlation offers little comfort if the number jumps toward 1 during sell-offs.


Diversifier, hedge, safe haven: three different jobs

These terms get used interchangeably. They shouldn't be.

  • A diversifier has returns that aren't closely tied to the rest of the portfolio over time.

  • A hedge is expected to move against a specific exposure.

  • A safe haven is expected to hold its value, or gain, during market stress.

Bitcoin can qualify as a diversifier over a ten-year backtest and still fail the safe-haven test in a particular crash. BlackRock's case rests on the first definition. Most of the skepticism, including JPMorgan's, is about the second.


How much of a portfolio should be in Bitcoin? Start with position size

There's no universal percentage, and anyone who offers one is skipping the part that matters. What you can do is see what a given weight implies.

Take a $100,000 portfolio. Suppose Bitcoin falls 50% and nothing else moves, with no trading, fees or taxes.

Starting BTC weight

Loss on the BTC position

Loss on the whole portfolio

2%

$1,000

1%

10%

$5,000

5%

30%

$15,000

15%

Arithmetic only. In a real sell-off, other holdings may fall at the same time, making the total worse.

The math is trivial, but people underestimate how fast weights move. If a 2% position doubles while everything else stays flat, Bitcoin is now about 3.9% of a $102,000 portfolio. After a strong year it can be several times its intended size, and per the table above, its share of total risk grows even faster than its weight. Knowing your current allocation, not the one you set up a year ago, is the basic requirement.


A note on diversifying within crypto

Everything above concerns Bitcoin's role next to stocks and bonds. Spreading money across many tokens is a separate question, and often a weaker one than it looks: many altcoins move closely with Bitcoin, especially in drawdowns. We cover that side in Crypto Diversification: Genuine Strategy vs. Illusion.


What to measure in your own portfolio

The institutional debate is useful because it points to the numbers that matter. For an individual portfolio, they are:

  1. Real exposure. Bitcoin and other crypto across every wallet and exchange, as a share of crypto holdings and as a share of total wealth.

  2. Current weights versus target. How far each position has drifted since you last rebalanced.

  3. Correlation over more than one window. Between your holdings, and against the markets you also invest in.

  4. Performance that separates skill from cash flows. Time-weighted return (TWRR) shows how the holdings performed. Money-weighted return (MWRR) shows how you did, including the timing of deposits and withdrawals. A large gap between the two usually means timing helped or hurt more than the assets did.

  5. Losses in currency, not just percentages. What a 40% or 60% drawdown means in dollars or euros.

The hard part is usually not the math but assembling the data when holdings are spread across several exchanges and self-custody wallets. That's what BitDiver Portfolio is built for: it connects supported wallets through public keys and exchanges through read-only API keys, then reconstructs your wealth history and calculates TWRR, MWRR, risk indicators and a correlation matrix across your crypto holdings. It covers the crypto side of the picture. Equities, bonds and other assets still need to be part of your overall allocation decision. For a deeper walk through these metrics, see our guide to crypto portfolio analysis.


FAQ

Is Bitcoin good for diversification?

It has been over some long periods, including the 2016–2026 window BlackRock studied. But its correlation with stocks changes over time and has risen during stress episodes, which is why firms like JPMorgan hesitate to treat it as a reliable diversifier.

There's no single figure. It depends on the equity index, the return frequency and the time window. Rolling correlations show it moving over a wide range rather than settling at one level.

That depends on your objectives, time horizon and capacity for loss, and this article isn't personal advice. A practical starting point is to calculate what a 50% drop would cost at your current weight, and how much of your portfolio's total risk that position represents.

Only if its correlation with the rest of the portfolio stays low. With high correlation, a 2% position can account for several times its weight in total portfolio risk, as the table above shows.


This article is for educational purposes only and is not personalized investment advice. Cryptoassets can lose substantial value. Historical and hypothetical results do not predict future performance.


Sources

  • J.P. Morgan Asset Management, 2027 Long-Term Capital Market Assumptions (October 2026)

  • BlackRock, Re-Underwriting Bitcoin (August 2026), pp. 2 and 12

  • BitDiver Portfolio, product features

  • BitDiver, Crypto Portfolio Analysis

  • News context: Criptovaluta.it, October 19, 2026

 
 
bottom of page