When to Use Bonds, When to Use Cash, and When a Short-Duration ETF Beats Both

Most investors learned the hard way in 2022 that “bonds” is not a single thing. The Bloomberg U.S. Aggregate Bond Index lost 13.0% that year, while the Federal Reserve took the policy rate from 0.25% to 4.50% in eleven months ^[1][2]. That was enough to turn a supposedly boring sleeve into a source of regret.

The better question is not whether to own bonds or cash. It is which kind of shortfall you can tolerate: price volatility, reinvestment risk, or the quiet drag of sitting in cash while short Treasuries pay more. A 3-month T-bill, a 1–3 year Treasury ETF like SHY, an aggregate bond fund like AGG, and a high-yield savings account all solve different problems. They also fail in different ways.

The 2022 lesson: duration, not just yield, did the damage

Bond investors often stare at yield and ignore duration. That is backward. Duration tells you how hard a bond’s price should move when rates change. Macaulay duration is the weighted average time to receive cash flows; modified duration translates that into an approximate price sensitivity. A fund with a modified duration of 6 years should lose about 6% for a 1 percentage-point rise in yields, before convexity and reinvestment effects ^[8].

That math explains why 2022 hurt. The Fed raised the target range from 0.00%–0.25% to 4.25%–4.50%, and the 10-year Treasury yield rose sharply as well ^[2]. Aggregate bond funds were exposed to both higher rates and longer duration. Short Treasuries were not. Cash was not. The difference was not subtle.

Here is the part many investors miss: yield is not return. Yield is a starting point, not a promise. If you buy a longer-duration bond fund and rates rise, the price drop can overwhelm the income for a long time. That is why a bond sleeve should be built around the job you need it to do, not the headline yield you want to see.

For a broader framework on measuring portfolio risk, see risk measurement and the three numbers that matter. Bond duration is one of those numbers. Ignoring it is expensive.

Source note: duration figures vary over time with portfolio composition; check the fund fact sheet before buying.

Three rate scenarios tell you more than one yield quote

Comparing cash, T-bills, SHY, and AGG requires a scenario, not a snapshot. The same yield can mean very different things depending on whether the Fed is still hiking, has paused, or is cutting. The table below uses simple assumptions to show the direction of travel, not a forecast. It is illustrative, not audited performance.

Assume a 1-year horizon, no taxes, and reinvestment at the prevailing short rate. For SHY and AGG, price impact is approximated using duration math; for cash and savings, the main risk is reinvestment. That is the tradeoff. Cash is stable today, but it can be slow to reprice if rates fall. Short Treasuries can capture higher yields with modest price risk. Longer bond funds can do well in cuts, but they can also get hit hard before the cuts arrive.

That table hides a blunt truth: if you think the Fed is near the end of a hiking cycle, short duration is usually the cleaner bet. If you think cuts are coming soon and you can tolerate volatility, longer duration has more upside. If you do not have a view, cash is the least wrong answer — but it is rarely the highest-return answer.

For readers who want a framework for identifying regime shifts, regime detection is the right lens. Bond allocation is mostly a regime call in disguise.

Cash is not “safe” in real terms. It is only stable in nominal terms. If inflation runs above your deposit rate, your purchasing power is leaking while the balance looks unchanged.

Why short-duration bonds often beat cash in hiking cycles

Short-duration bond funds tend to shine when policy rates are rising or staying elevated. The reason is simple. They reset faster than cash accounts at many banks, and they usually offer a cleaner pass-through of Treasury yields than retail deposit pricing. The Federal Reserve’s H.15 data show that short Treasury yields and money-market rates moved up quickly during the 2022–2023 hiking cycle, while bank deposit rates adjusted more slowly ^[5].

That lag matters. Banks do not reprice deposits instantly, especially when they already have plenty of cheap funding. A high-yield savings account can be excellent for emergency money, but it is not always the best place to park six or twelve months of idle cash if 3-month bills are paying more. The spread can be real. It is not theoretical.

Morningstar’s fund-flow data showed record inflows into ultrashort bond funds in 2023, as investors chased yield with less duration risk than traditional bond funds ^[7]. That was a rational response to the market. It was also a confession: many investors wanted something that behaved more like cash but paid more like bonds.

That is where a short-duration ETF can beat both. It can offer higher yield than a bank account, lower volatility than AGG, and less reinvestment friction than rolling individual bills if you want a one-ticket solution. The catch is that it is still a bond fund. If rates fall quickly, the yield resets down. If credit risk creeps in, the fund can underperform Treasuries. Read the fact sheet, not the ticker symbol. For a quick refresher on fund disclosures, see how to read a fund fact sheet.

AGG is not broken; it is just built for a different job

Aggregate bond funds are often treated as the default “safe” sleeve. That is lazy portfolio construction. AGG owns a broad mix of Treasuries, agency MBS, and investment-grade credit, so it is designed to diversify equity risk over full cycles, not to act like a parking lot for money you may need soon ^[4].

AQR’s defensive-equity research has long argued that low-risk assets can be useful, but only when investors understand the source of protection. Bonds help because they often rally when growth slows and policy eases. They do not help when inflation shocks push yields higher at the same time stocks are falling ^[9]. That was the 2022 problem in one sentence. Stocks and bonds both got hit.

Most investors overread the word “defensive.” Defensive does not mean immune. It means the asset tends to behave differently from stocks in some regimes. In a disinflationary slowdown, AGG can be excellent. In an inflation shock, it can be a disappointment. If you need ballast for a stock-heavy portfolio and you can tolerate interim volatility, AGG still has a role. If you need a place to wait six months for a house down payment, it is the wrong tool.

This is where the distinction between total return and utility matters. A bond fund can be a good diversifier and a bad cash substitute at the same time. That sounds contradictory only if you think every portfolio sleeve should do one job. It should not.

For investors comparing bond sleeves inside a broader allocation, the 60/40 debate and stocks vs. bonds vs. cash are the right companion pieces.

AGG is a portfolio diversifier first and a cash substitute second. Treating it like a savings account is how investors rediscover duration the hard way.

A worked example: $100,000 parked for 12 months

Suppose you have $100,000 you may need in a year. You are choosing among a 3-month T-bill ladder, SHY, AGG, and a high-yield savings account. The right answer depends on your rate view, but the wrong answer is easier to spot: anything with meaningful duration risk if the money is truly earmarked.

Here is a simplified walkthrough. If the Fed stays roughly where it is, a T-bill ladder and SHY should both deliver something close to prevailing short rates, minus small expenses. A high-yield savings account may trail if the bank keeps part of the spread. AGG could do fine, but it is taking more price risk for no guarantee of extra return. If the Fed cuts sharply, AGG likely wins on price appreciation. If the Fed hikes again, the short-duration choices recover faster.

That is the uncomfortable implication. The “best” choice depends on whether you are investing or waiting. Waiting money should not be forced into a long-duration sleeve just because the yield screen looks attractive. Investing money can tolerate more duration because the holding period is longer and the price swings are part of the bargain.

Use this checklist before you choose:

1. Do I need the money inside 12 months?
2. Can I tolerate a 2%–5% interim drawdown?
3. Do I want the highest current yield, or the best chance of preserving purchasing power?
4. Am I comparing after-tax yields, not just headline yields?

If you want a broader framework for deciding how much risk belongs in a portfolio sleeve, an investment policy statement is more useful than a gut feeling.

The historical record favors short duration in hikes and longer duration in cuts

History is not a guarantee, but it is a decent map. During hiking cycles, short-duration bonds usually hold up better than intermediate or long-duration funds because their prices are less sensitive to rising yields. During cutting cycles, the opposite is often true: longer duration gets a bigger price boost when yields fall. That pattern is visible in Treasury return data and in fund behavior across cycles ^[10].

The key is that the cycle matters more than the label. “Bond” is too broad. A 2-year Treasury and a 10-year Treasury are not cousins; they are different animals. The 2-year is mostly a carry trade with limited price sensitivity. The 10-year is a macro bet on growth, inflation, and policy. If you do not want to make that bet, do not pretend you are not making it.

One more wrinkle: reinvestment risk. Short-duration assets force you to roll over at whatever rates are available later. That is great when rates are rising and painful when they are falling. Long-duration bonds lock in more yield today, but they expose you to bigger price swings. There is no free lunch. There is only a choice about where you want the risk to live.

For readers who want to think more systematically about changing market conditions, regime detection and Sharpe vs. Calmar are useful complements. A bond sleeve that looks good on Sharpe can still be miserable if the drawdown arrives at the wrong time.

Short duration is not a permanent winner. It is a regime tool. Use it like one.

A decision tree for the bond sleeve you actually need

Start with the money’s purpose, not the product. If the cash is for an emergency fund, a tax bill, tuition, or a house down payment, principal stability comes first. High-yield savings or a T-bill ladder usually makes more sense than a bond fund. If the money is part of a long-term portfolio and you want equity ballast, AGG or a similar intermediate bond fund can still earn its keep.

Then ask the rate question. If you think policy rates are near a peak and you want yield without much volatility, short-duration Treasuries are often the sweet spot. If you think cuts are coming and you can tolerate mark-to-market swings, longer duration has more upside. If you have no view, cash is fine — but do not confuse “fine” with “optimal.”

Here is the decision tree in plain English:

• Need the money within 12 months? Use cash or T-bills.
• Need a low-volatility yield sleeve? Favor short-duration Treasury ETFs.
• Need diversification against equity drawdowns over years? AGG still has a role.
• Expect rapid rate cuts? Longer duration may outperform, but only if you can stomach interim losses.

That is the real choice set. Everything else is packaging.

For investors who want to compare implementation vehicles, ETFs vs. mutual funds and tax-aware rebalancing matter more than most people think. A good idea can still be a bad trade if the wrapper is wrong.