Jack Bogle's asset allocation and Vanguard's balanced funds

Jack Bogle, who founded Vanguard in 1974, said in a Spring 2017 interview that he was "about 50/50 in bonds and stocks", and CNBC had reported in 2015 that his portfolio was entirely in US markets.

Tested against MSCI World from January 2001 to September 2026, the two Vanguard funds tied to him, the Balanced Index Fund (VBINX) and Wellington Fund (VWELX), were partly set up for different market scenarios and cushioned equity crashes without protecting against them. Their volatility-adjusted returns were partly better: ahead over 20 years and the full window, not over the last 10. Net of fees VBINX's returns were not better in any window, and Wellington's were partly better, ahead over the full window alone.

Key takeaways

1. Set up for different market scenarios: partly. Both funds lost less than MSCI World in all four falls we tested, but VBINX almost as much in the 2022 rate shock, and both gained about 62% in the 2023 to 2026 rally against 98.8%.
2. Guard against catastrophic equity falls: cushions, does not protect. About a third in bonds held their worst fall to 32.6% (VBINX) and 32.5% (Wellington) in 2007 to 2009, against 54.0% for MSCI World, but nothing in either fund pays out when stocks crash.
3. Better volatility-adjusted returns than MSCI World: partly. VBINX had a higher Sharpe ratio than MSCI World in 181 of 190 rolling 10-year windows and Wellington in 189, yet over the 10 years to September 2026 VBINX trailed and Wellington was level.
4. Better returns net of fees than MSCI World: no for VBINX, partly for Wellington. VBINX compounded less than MSCI World over 10 years, 20 years and the full window; Wellington beat it only over the full window, 8.3% a year against 7.5%, and was close over 20 years.
5. Bogle's own mix was simple and leaned towards bonds with age. Reports put him at 60/40 for years and 50/50 by 2015, all in US stocks and bonds; he called his age-in-bonds rule "not a formula".

How do Bogle's balanced funds score on our four tests?

We judge both funds, net of their expenses, against MSCI World Net in US dollars, with one-month Treasury bills as cash, using monthly returns to September 2026. They give the same answer on the first three tests and differ on the fourth.

Windows: 10 years, Oct 2016 to Sep 2026; 20 years, Oct 2006 to Sep 2026; full window, Jan 2001 to Sep 2026. MSCI World is a developed-market equity index, not a balanced benchmark.

What was Jack Bogle's own asset allocation?

Bogle's dated statements describe a simple portfolio of US stock and US bond funds whose bond share rose with age. They are listed by date because they are not fully consistent.

The 2010 and 2015 accounts cannot both describe the same total portfolio, and none gives exact holdings. Explaining the move to 50/50, Bogle told CNBC: "I just like the idea of having an anchor to the windward. I'm not so much worried about having my estate grow." He died on 16 January 2019, aged 89, according to Vanguard's announcement.

What was Bogle's rule of thumb for stocks and bonds?

Bogle suggested a bond position roughly equal to your age, and repeatedly said it was a starting point rather than a formula. In a March 2014 Motley Fool interview he said: "your bond position should equal your age -- but that's a rule of thumb".

In the 2017 interview he went further: "I'm afraid it's a bad rule of thumb because people forget to add Social Security to their bond position." He valued a typical retiree's Social Security at around $300,000, so his rule measures bonds against the whole balance sheet, including pension income, not the investment account alone.

Did Bogle own international stocks or the three-fund portfolio?

No: Bogle held only US stocks and US bonds, and the Bogleheads three-fund portfolio comes from Taylor Larimore, not from Bogle's own holdings. "I continue to believe the U.S. is the best place to invest and, therefore, I don't think you need to own non-U.S. stock," he said in the 2017 interview (p. 28). NPR's 2008 summary records a suggested maximum of 20% of the stock allocation for anyone who wanted them.

Larimore's 2018 book, according to Wiley's July 2018 announcement, invests in three index funds: US stocks, non-US stocks and US bonds. In his foreword, as quoted by The White Coat Investor, Bogle called Larimore's 20% international share "a compromise between my suggested maximum of 20% and the minimum 20% recommended by a Vanguard study." After noting that the S&P 500 had risen 743% since 1994 against 237% for the EAFE index of non-US stocks, he wrote: "That I was right is beside the point. It may have been luck."

What do Vanguard's Balanced Index and Wellington funds hold?

At 30 September 2026, the Balanced Index Fund held 61.6% stocks and Wellington Fund 66.4%, according to Vanguard. They are the funds most closely tied to Bogle: one he named as a single-fund portfolio, the other the balanced fund where his career began.

At 30 September 2026: Balanced Index (VBINX) 61.64% stocks, 37.82% bonds, 0.54% short-term reserves; Wellington (VWELX) 66.39% stocks, 32.73% bonds, 0.88% short-term reserves. Source: Vanguard fund data for VBINX and VWELX; chart data (CSV). Each fund's shares are of its own assets and sum to 100%.

Balanced Index Fund. Its April 2026 summary prospectus says it tracks the CRSP US Total Market Index with about 60% of assets and the Bloomberg U.S. Aggregate Float Adjusted Index with about 40%. In the 2014 interview Bogle said you could invest with one fund, "something like the Vanguard Balanced Index Fund. It's 60% total stock market, 40% total bond market, both U.S." The Investor shares cost 0.18% a year and are marked as closed to new investors; the Admiral shares (VBIAX) cost 0.07%.

Wellington Fund. Walter Morgan founded it in 1929 and hired Bogle in 1951, according to Vanguard's 2019 announcement; Bogle created Vanguard in 1974 after a dispute at Wellington Management. That firm still runs the fund actively, with 60% to 70% in "dividend-paying and, to a lesser extent, non-dividend-paying" stocks of established large companies and the rest mainly in investment-grade bonds, per its March 2026 summary prospectus. Investor shares cost 0.24%, and the stock portfolio held 85 names at 30 September 2026.

Other investors in this series kept their own money differently from their funds; see Warren Buffett's 90/10 instruction and David Swensen's index-fund mix for individuals.

Are Bogle's balanced funds set up for different market scenarios?

Partly: both behaved like a scaled-down equity portfolio rather than one built for distinct regimes, falling less in every crash we tested and rising less in rallies.

Cumulative returns in USD, VBINX / Wellington / MSCI World: dotcom (Jan 2001 to Sep 2002) -16.6% / -8.6% / -38.1%; financial crisis (Nov 2007 to Feb 2009) -32.6% / -32.5% / -54.0%; Covid (Feb to Mar 2020) -12.3% / -14.1% / -20.6%; 2022 -17.0% / -14.3% / -18.1%; rally (Jan 2023 to Sep 2026) +62.2% / +61.8% / +98.8%.

Source: our calculations from month-end total returns (VBINX and MSCI World monthly data, VWELX monthly data); chart data (CSV). Funds are net of expenses, the index carries no costs, and the episodes are fixed calendar windows.

In the 43 months when MSCI World rose 5% or more, VBINX gained 4.5% on average and Wellington 4.6%, against 7.2% for the index. In the 29 months when it fell 5% or more, both lost 4.9% on average against 8.6%.

The exception was 2022, when inflation pushed up interest rates and bonds fell with stocks: VBINX lost 17.0% against 18.1% for MSCI World (why bonds protected in 2008 but failed in 2022). Wellington's smaller dotcom loss, -8.6% against VBINX's -16.6%, came from an actively managed portfolio, so it reflects stock selection as well as the balance of stocks and bonds.

Do Bogle's balanced funds guard against catastrophic equity falls at all?

They cushion falls, mainly through roughly a third in investment-grade bonds; nothing in either fund protects with a contractual payoff such as a put option. VBINX's worst month-end fall was 32.6%, from October 2007 to February 2009, and it was back at its previous peak by December 2010; Wellington fell 32.5% and recovered at the same time. MSCI World fell 54.0%, a loss that needs a 117% gain to recover, and took until July 2013.

The cushion has limits. VBINX's monthly returns had a 0.97 correlation with MSCI World, so about 93% of their variation lined up with the equity index (our calculation), and both funds fell in every one of the 29 months when MSCI World lost 5% or more. Capital split 60/40 does not split swings 60/40, as our analysis of how a 60/40 split divides volatility shows.

Holding thousands of stocks does not change that. At 30 September 2026, the ten largest of VBINX's 3,174 stocks made up 34.8% of its listed stock value, with NVIDIA alone at 7.2% (our calculation from Vanguard's holdings file). MSCI World's factsheet for the same date, 30 September 2026, showed 1,249 constituents, 27.85% in the top ten and 72.94% in the United States. The 30 October 2026 edition, now at the same link, shows 1,248 constituents and 72.77% in the United States.

Wellington is more concentrated by design: 85 stocks, with the top ten at 46.1% of stock value. Weights that follow prices spread holdings, not drivers; see our analysis of asset class correlations.

Are Bogle's balanced funds' volatility-adjusted returns better than MSCI World?

Partly: better over 20 years, the full window and almost every rolling decade, but not over the 10 years to September 2026, when VBINX trailed and Wellington was level. The Sharpe ratio divides the average monthly return above cash by the volatility of that excess return, annualised. The Sortino ratio divides the same excess return by downside deviation, so only months that fell short of cash enter the denominator.

Volatility is not the same as the chance of a permanent loss, so these ratios describe how smoothly each fund earned its return, not how safe it was. A like-for-like check holds volatility equal: MSCI World diluted with 37.3% T-bills to VBINX's 9.6% volatility would have compounded 5.6% a year since 2001 with a 37.3% worst fall, against VBINX's 7.1% and 32.6%.

Both funds were chosen because they survived and are linked to Bogle. The ratios also move with the cash benchmark; see Sharpe ratio versus CAGR for why a higher ratio does not by itself mean more wealth.

Are Bogle's balanced funds' returns net of fees better than MSCI World?

For VBINX, no in every window; for Wellington, partly: better over the full window since 2001, close over 20 years and worse over 10. Fund returns are net of fees; MSCI World Net is an index and carries no costs, while a fund that tracks it does (the iShares Core MSCI World UCITS ETF charges 0.20% a year), so a shortfall smaller than such a fee counts as close rather than a clear loss.

Wellington trailed over 20 years by less than 0.1 points a year, so that window is close; VBINX's shortfalls, 0.3 to 3.6 points a year, are not. Fund figures are net of expenses of 0.18% (VBINX Investor shares) and 0.24% (Wellington Investor shares).

Cost was Bogle's main lever on compounded growth: in 2017 he said that if the market returns 4% and active funds take 2% in costs, "that's a net return of 2%." As a hypothetical illustration, deducting an extra 0.64% a year (the gap between Vanguard's cited 0.82% average for moderate allocation funds and VBINX's 0.18%) from VBINX's monthly returns cuts $10,000 grown since January 2001 from $58,837 to about $49,900.

Home bias helped both funds against the index. From January 2001 to September 2026, MSCI USA compounded 8.5% a year and MSCI World ex USA 5.8%, both net in USD (our calculation from MSCI index levels), while the US share of MSCI World rose from about half in 2001 (our estimate) to 72.94% in September 2026. That gap is a market outcome, not a feature of the 60/40 split, as Bogle's own "may have been luck" acknowledges.

Weighing the four results

The two funds gave up a large part of equity rallies, more than a third of MSCI World's 2023 to 2026 gain, in exchange for falls about 40% smaller in the 2008 crash, more than half smaller in the dotcom crash, and a quicker recovery.

That trade earned a better return per unit of volatility over most decades, but not more money for VBINX, and the cushion thinned in 2022 when bonds fell too. Whether that exchange suits a particular investor depends on their spending needs and how much of a fall they could sit through, which these figures cannot settle.

What can an individual investor learn from Jack Bogle?

Bogle's most durable lesson is that costs and behaviour are the parts of investing an individual controls; his own allocation was simple, home-biased and set by age and temperament. Questions to ask of your own portfolio:

• What do you pay in total fund costs each year, and what would that compound to over your horizon?
• If you count pension income as a bond-like asset, how much of your whole balance sheet depends on equities?
• What in your portfolio would hold its value if stocks and bonds fell together, as in 2022?
• How much of your equity sits in the ten largest companies once you look through your index funds?
• Could you hold your mix through a 33% fall without selling, as VBINX investors faced from 2007 to 2009?

Our conservative asset allocation guide compares balanced mixes with other ways of limiting losses, and Cliff Asness's approach to diversification offers a contrasting view of the 60/40 portfolio. For an active manager's record, and the gap between a fund's return and what its investors earned, see Peter Lynch's Magellan strategy.

How can copio help you examine your own allocation?

Copio is in private beta across Europe, with early access by application. It looks through your funds into the holdings and asset classes underneath, and its concentration analysis shows what a modelled fall in your largest position would do to the whole portfolio.

A return-correlation map places each asset class as a main driver, a true diversifier, dead weight or a shock absorber, so you can see whether a 60/40 split spreads the swings or only the capital.

A crash test shows how a mix like yours would have fared in past crises, and an efficient frontier plots your portfolio against a modelled frontier of assumed growth and volatility, labelled as not a forecast. Copio provides educational analysis, not personalised advice, and does not place trades or find an optimal allocation for you. The free portfolio checker is a first step.

For money already available to invest, our lump-sum versus phased-investing study compares the cost of waiting with losses after badly timed starts, including record-high entry dates.