Open the analyzer →

Home › Explore Investing Strategies › Asset allocation and portfolios

Asset allocation: cash 30, stocks 70, bonds and rebalancing

Stocks with cash, IEF or TLT since August 2002: return, max drawdown, recovery time, cash interest, 2008 vs. 2022, and monthly, yearly or no rebalancing.

In one line

A 30% cash share cut SPY's max drawdown from −55.2% to −42.4% and lowered the annual return from 11.2% to 8.0%. Holding 30% in intermediate Treasuries instead returned 9.1% a year with a −39.0% max drawdown, but in 2022 the bonds fell with stocks. A 70/30 that was never reset ended 24 years later with 93% in stocks.

Backtest results for this strategy: How much less does a 70% stock, 30% cash mix fall than 100% stocks?

Article

Most of the losses individual investors take come not from picking the wrong stock but from putting too much in at once. Two people who bought the same stock, one with 10% of their assets and one with 90%, live through the same 30% drop very differently. For the first it is a bearable correction; for the second it turns every decision into whether to sell now. That is why how much to put in comes before what to buy.

A 30% cash share looks like a drag on returns, but it does three jobs: it lets you buy in a crash (opportunity), keeps one bad pick from sinking the whole account (defense), and above all reduces impatience (psychology). The numbers below show how much return that cost.

Asset allocation is the rule that decides how much goes into stocks, bonds and cash before any stock is picked. This article calculates with SPY and US Treasury ETFs what a rule like cash 30, stocks 70 actually changed. It also covers interest on cash, bonds in place of cash, and different rebalancing intervals. Figures are as of September 2026 and were calculated by the operator.

How much of the result allocation decides

In a 1986 Financial Analysts Journal paper, Gary Brinson, Randolph Hood and Gilbert Beebower analyzed the quarterly returns of 91 large US pension plans from 1974 to 1983 and reported that investment policy, the planned allocation, explained 93.6% of the variation in returns over time. The number is often misquoted as "allocation decides 93.6% of returns". In 2000, in the same journal, Roger Ibbotson and Paul Kaplan split the question in three. Using 10 years of monthly returns for 94 US balanced funds and 5 years of quarterly returns for 58 pension funds, policy explained about 90% of a fund's ups and downs over time, about 40% of the differences between funds, and on average about 100% of the return level. Allocation largely decides how much an account swings, and differences between people with the same allocation also come from stock selection, costs and timing.

Test conditions

$10,000 went in once on August 1, 2002, with nothing added later. Stocks are SPY, and bonds are the US intermediate Treasury ETF IEF (7 to 10-year maturities) and the long Treasury ETF TLT (20+ years), all with dividends reinvested. The target weights were bought at the first day's open and reset at the open of the first trading day of each month, with a 0.15% cost on each trade. Cash earned no interest, except in one row where it earned the US 3-month Treasury bill rate (^IRX). The period ends on September 30, 2026. The stock and cash rows match this site's backtest engine, and the operator calculated the bond rows separately with the same rules. Results of investing every month while holding the weights are in the stock and cash mix backtest.

Results (as of September 2026)

CaseEnd balanceAnnual returnMax drawdownTime to regain high20082022
SPY 100%$129,95711.2%−55.2%4.9 years−36.8%−18.2%
Stocks 90, cash 10$103,15510.1%−51.2%4.4 years−33.6%−16.3%
Stocks 70, cash 30$63,8888.0%−42.4%4.4 years−26.9%−12.6%
Stocks 70, cash 30 (T-bill interest)$72,1118.5%−42.0%4.4 years−26.6%−12.1%
Stocks 50, cash 50$38,6815.8%−32.2%4.4 years−19.8%−9.0%
Stocks 70, IEF 30$82,1419.1%−39.0%3.3 years−23.0%−16.9%
Stocks 70, TLT 30$85,0989.3%−38.0%3.3 years−19.7%−22.0%
Stocks 60, IEF 40$69,4888.4%−32.6%2.8 years−17.9%−16.6%

Time to regain high is how long it took to get back to the prior high after the 2007 to 2009 decline, where every max drawdown occurred. The 2008 and 2022 columns are calendar-year returns.

What cash did

A 30% cash share cut the max drawdown from −55.2% to −42.4%, by 12.8 points, and in exchange lowered the annual return from 11.2% to 8.0%, by 3.2 points. The time to regain the high shrank only from 4.9 years to 4.4 years, by five months. Cash made the decline shallower, but what lifted the account off the bottom was the remaining stock share. With the US 3-month Treasury bill rate paid on cash, the annual return was 8.5%. That rate averaged 1.71% a year over the period, and was close to zero at 0.07% on average from 2009 to 2015 and 0.06% from April 2020 through 2021.

The monthly reset rule used cash to buy stocks during the decline. The stock share of the 70/30 cash account never fell below 63.0% even in the 2008 decline, and went back to 70% on the first trading day of each following month. The saying that cash "lets you buy in a crash" shows up in the numbers only when a rule like this is set in advance.

2008 and 2022: what bonds did

Holding 30% in the intermediate Treasury fund IEF instead of cash gave 9.1% a year and a −39.0% max drawdown, better than cash on both counts, and regained the 2007 high in 3.3 years, 14 months sooner than cash. A 30% share of long Treasuries (TLT) was similar at 9.3% a year and −38.0%. The difference was decided in the two years below.

Calendar-year returns in 2008 and 2022. In 2008 SPY returned −36.8%, IEF +17.9% and TLT +34.0%, and stocks 70, TLT 30 returned −19.7%. In 2022 SPY returned −18.2%, IEF −15.2% and TLT −31.2%, and stocks 70, TLT 30 returned −22.0%, a larger loss than SPY. ① 2008 and 2022: the same 30%, different results Calendar-year returns with dividends. Mixes reset their weights at each month start. 2008 2022 SPY −36.8% −18.2% IEF (7 to 10-year Treasuries) +17.9% −15.2% TLT (20+ year Treasuries) +34.0% −31.2% Stocks 70, cash 30 −26.9% −12.6% Stocks 70, IEF 30 −23.0% −16.9% Stocks 70, TLT 30 −19.7% −22.0% Stocks 60, IEF 40 −17.9% −16.6% In 2008 Treasuries rose and softened the stock loss. In 2022 they fell with stocks. Correlation of SPY and IEF daily returns: −0.40 from Aug 2002 to 2021, +0.11 from 2022 to Sep 2026

In 2008 SPY returned −36.8%, while IEF returned +17.9% and TLT +34.0%. Stocks 70, TLT 30 lost 19.7%, a little over half of SPY's loss. 2022 was the opposite. While the US 3-month Treasury bill rate rose from 0.05% at the start of the year to 4.26% at the end, SPY returned −18.2%, IEF −15.2% and TLT −31.2%, and stocks 70, TLT 30 fell 22.0%, more than SPY. The correlation of SPY and IEF daily returns was −0.40 from August 2002 through 2021 and +0.11 from 2022 to September 2026. Since 2003, SPY fell in 2008, 2018 and 2022; IEF rose in 2008 (+17.9%) and 2018 (+1.0%) but fell with stocks in 2022, the year rates rose quickly. The longer the maturity, the more a bond moves with rates: TLT fell 48.4% from its August 4, 2020 high to October 19, 2023 and had not regained that high by September 2026. IEF fell 23.9% over the same span.

Rebalancing: monthly, yearly or never

Stock share of an account that started in August 2002 at 70% SPY and 30% IEF. Without rebalancing it fell to 56.1% on March 9, 2009 and reached 93.2% in September 2026. Rebalanced on the first trading day of each year it stayed between 52.3% and 76.3%, and monthly between 62.9% and 73.4%. ② The 70/30 left alone: 93% stocks after 24 years Stock share (month-end) of an account started in Aug 2002 at 70% SPY, 30% IEF 50% 60% 70% 80% 90% 100% 2005 2010 2015 2020 2025 Mar 9, 2009 56.1% Sep 2026 93.2% Never rebalanced Reset each January Target 70% The account reset every month stayed between 62.9% and 73.4%.

The stocks 70, IEF 30 account returned 9.1% a year when reset monthly, 9.2% when reset on the first trading day of each year, and 9.9% when left as bought. Max drawdowns were close at −39.0%, −36.7% and −38.2%, with 576, 50 and 2 fills and $341, $212 and $162 in costs. The account left alone earned the most because stocks rose far more than bonds and the stock share kept growing. The stock share, 56.1% on March 9, 2009, rose to 93.2% by September 2026. It is still called 70/30, but it will meet the next decline as a 93% stock account.

The monthly reset account's 9.1% a year was 0.3 points above the 8.8% from weighting the two assets' annual returns 70 to 30. That extra came from trimming what rose and buying what fell, the rebalancing bonus William Bernstein described in a 1996 essay (more in Grid trading). Over this period, though, letting the stock share grow earned more. In a 2010 Vanguard paper, Colleen Jaconetti, Francis Kinniry and Yan Zilbering concluded that the goal of rebalancing is to keep the chosen risk, not to raise returns, that risk-adjusted returns were not meaningfully different with monthly, quarterly or annual rebalancing, and that more frequent rebalancing mainly added cost. The numbers above looked the same.

When the rule holds and when it does not

  • It held for reducing drawdowns: every mix we calculated had a smaller max drawdown than SPY. For the same 30%, Treasuries cut the 2008 decline and the recovery time more than cash did.
  • It protects less when stocks and bonds fall together: in 2022 stocks 70, IEF 30 lost 16.9%, more than stocks 70, cash 30 (−12.6%).
  • It earns less in rising markets: since August 2002, SPY 100% ended at $129,957 and stocks 70, cash 30 at $63,888. That gap was the price of the smaller drawdown.
  • Left alone, it becomes a different account: without rebalancing, the asset that rose most takes a growing share, and the account no longer carries the risk you chose. An account started at stocks 70, cash 30 and left alone was 96.8% stocks in September 2026.

Criteria for setting the weights

Three criteria are commonly fixed before any numeric weight. The calculations in this article cover only the first.

  • Market regime: raise the stock share in a long uptrend and hold more cash in a long downtrend. How this site judges the regime is in the methodology.
  • Cap per stock: keep any single stock below a set share of the whole (for example 20%). The more confident you are, the more you need this rule.
  • Loss tolerance: size the position so that a stop-out costs 1 to 2% of total assets. The distance between entry and stop sets the quantity (Plan before entry).

Instead of a fixed schedule, you can also rebalance by deviation: reset only when a weight drifts more than a set band (for example 10 points) from the target. That method is not included in these calculations.

Related concepts and search terms

  • Strategic and tactical asset allocation: keeping long-term target weights, versus changing weights with market conditions. This article's calculations are the first kind. An example that changes leverage by regime is Leverage cross-rebalancing.
  • 60/40 portfolio: the classic mix of 60% stocks and 40% bonds. The stocks 60, IEF 40 row in the table is one.
  • Rebalancing bands: resetting only when a weight drifts beyond a set width. The Vanguard paper gives annual or semiannual checks with rebalancing at 5-point drifts as a reasonable example.
  • Duration: roughly how many percent a bond's price moves when rates move one point. It is larger for longer maturities, which is why TLT rose more than IEF in 2008 and fell more in 2022.
  • Permanent portfolio, risk parity: methods that mix assets that do well in different economic conditions, or size weights so each asset carries similar risk. Harry Browne's permanent portfolio holds 25% each in stocks, long Treasuries, gold and cash.

Search terms: asset allocation, 60/40 portfolio, rebalancing frequency, rebalancing bonus, stock bond correlation 2022, Brinson Hood Beebower, cash drag, Ibbotson Kaplan 40 90 100.

Drawbacks and risks

  • The calculation covers one period, August 2002 to September 2026. IEF and TLT had not regained their August 2020 highs by September 2026, and different start and end dates can change the bond results a lot.
  • The row with interest on cash uses the US 3-month Treasury bill rate as is. Real savings accounts and short-term bond ETFs earn different returns, and Korean investors also face currency and taxes.
  • Each sale for rebalancing can realize capital gains. For Korean residents, overseas stock gains above ₩2.5 million a year are taxed at 22% (including local income tax).
  • The weights are examples for calculation, not weights that fit everyone.

Operator's assessment

The numbers on asset allocation say two things. First, the smaller drawdowns were mostly paid for with return. A 30% cash share cut the max drawdown by 12.8 points and the annual return by 3.2 points. Second, what you mix in mattered as much as the weight. The same 30% in intermediate Treasuries rose in 2008 and regained the high 14 months sooner than cash, but fell with stocks in 2022 and did worse than cash. So we do not see Treasuries as always better than cash. They were different tools: cash was more useful in 2022, when rates rose quickly, and Treasuries in 2008, when stocks fell hard. Rebalancing was less a way to add return than a way to keep the risk chosen at the start, and the 70/30 never reset became a 93% stock account after 24 years. Confluence Zone does not set anyone's weights; it shows the regime and the state of each stock, and stops there.

What this article does not cover

This article does not set weights for your age, income, taxes or debt. That call belongs to a registered investment adviser, and the weights here are examples for calculation. The figures are past results calculated the same way as this site's engine (dividends reinvested, 0.15% cost per trade, taxes and currency left out). Only the period since August 2002, when US Treasury ETF data begin, is covered. Everything here is reference information based on past, public data and is not investment advice.

Articles and questions on the same topic

Monthly investing and lump sums

Asset allocation and rebalancing

Related

Stock and cash mix backtest · SPY monthly vs. lump sum · Grid trading · Leverage cross-rebalancing · Klarman's absolute-return margin of safety · Investor psychology

Basics and trading strategies

← Support, resistance and breakouts: 52-week highs and pullbacks · Contents · All Weather and the Permanent Portfolio: four-season allocations tested with ETFs →

See these metrics on a real stock →

Confluence Zone is an analysis tool for information and research; it is not investment advice, a solicitation, or a trading instruction. Investment decisions and their outcomes are your own responsibility.
About & methodology · Explore Investing Strategies · Glossary · Questions and answers · Rule status · Sample analysis · Pricing · Terms · Investment disclosure · Privacy policy · Open the analyzer · 한국어 · 日本語 · 繁體中文 · 简体中文