What Actually Counts as Risk-Free
Every portfolio model leans on a "risk-free rate" as its baseline, but almost nothing in investing is free of every kind of risk. Knowing exactly which instruments come closest, and under what conditions that safety quietly disappears, keeps you from treating any conservative-sounding investment as automatically safe.
What actually qualifies as risk-free
An asset earns the label "risk-free" in portfolio theory only if its return is known with certainty over the investor's specific holding period, no more and no less. Short-term U.S. Treasury bills come closest to satisfying that condition: they are backed by the taxing power of the federal government, which makes default risk effectively negligible for practical purposes, and they mature in a year or less, so an investor who holds one to maturity locks in a known return with essentially no price risk along the way. That is why the standard risk-free rate used throughout academic and practical finance is the current short-term Treasury bill yield, not the yield on any other government security.
Notice what this definition deliberately excludes. A 10-year Treasury bond carries the same government backing and therefore the same negligible default risk as a 3-month bill, but it is not risk-free for most investors, because its price can move substantially before maturity if interest rates change, and most investors are not planning to hold it for exactly ten years without ever needing the money sooner. Risk-free, in other words, is a statement about a specific instrument matched to a specific horizon, not a blanket property of "government bonds" as a category.
The math of maturity mismatch
The tool for measuring how much a bond's price moves when rates change is duration, and the approximation used across the industry is: percentage price change ≈ -duration × change in yield. Consider an investor who buys a 10-year Treasury bond with a duration of approximately 8.5 years, intending to use the money for a home down payment in exactly one year. If interest rates rise by 1.5 percentage points over that year, a plausible move in a single year given historical rate volatility, the bond's price falls by roughly 8.5 × 1.5% = 12.75%. An investor who needed a "safe" holding for a one-year goal and chose a 10-year bond instead of a short bill could show up at closing with 12.75% less than expected, despite having owned an asset with literally zero default risk the entire time.
Now run the comparison the other way. Suppose the same investor instead holds a 10-year time horizon but keeps rolling three-month Treasury bills, reinvesting every quarter at whatever the prevailing rate happens to be. Over the decade, if the average short-term rate turns out to be 3.5%, the ending real value depends heavily on the path rates took, but there is no price risk at any single roll date, only reinvestment rate risk, the uncertainty of what rate will be available next quarter. If rates instead average 5.5% over the decade due to a higher-inflation environment, the rolled bills end up ahead in nominal terms, but the investor bore genuine uncertainty about the outcome the entire time, just a different kind of uncertainty than a long bond's price risk. Neither instrument is free of every risk over a 10-year horizon; they simply carry different flavors of it.
A second worked example: laddering versus a single maturity
A practical middle ground between rolling three-month bills and holding a single long bond is a bond ladder, a set of holdings with staggered maturities that spreads reinvestment dates across time rather than concentrating them all at once. Consider an investor with a five-year horizon who splits $50,000 evenly across bonds maturing in one, two, three, four, and five years, $10,000 in each. If rates rise sharply in year one, only the $10,000 maturing that year gets reinvested at the new, higher rate immediately; the other four rungs continue earning their original locked-in rates until each one matures in turn. Compare that to an investor who put the full $50,000 into a single five-year bond: that investor is fully protected from the year-one rate move, in the sense that nothing needs reinvesting, but is also fully exposed to price risk if any of the $50,000 needs to be accessed before the five years are up, since the whole balance sits in one instrument whose market price moves with rates in the interim.
Neither structure is risk-free in an absolute sense, the ladder carries partial reinvestment risk each year, and the single bond carries full price risk until maturity, but the ladder smooths the exposure across time rather than concentrating it at a single reinvestment date or a single maturity date, which is precisely why laddering is a common technique for investors managing a known sequence of near-term cash needs, a series of tuition payments or a phased home purchase, for example, rather than a single lump sum need.
What the historical record shows
The gap between nominal safety and real safety has shown up repeatedly across market history. Treasury bills have historically delivered a nominal return that, averaged over long multi-decade stretches, sits only modestly above the average inflation rate over the same stretches, leaving a real return close to zero across the full historical record, even though the nominal principal never once declined. Investors who equate "the balance never goes down" with "I am not losing anything" are technically correct in nominal terms and frequently wrong in the terms that actually matter for long-term purchasing power.
Inflation-protected securities, which adjust their principal with a published price index, were designed specifically to close this gap for long-horizon safe money, offering a return that is closer to genuinely risk-free in real, purchasing-power terms rather than merely nominal terms. They are not perfect either: their market price still moves with real interest rates before maturity, and their tax treatment in a standard taxable account creates a well-known complication where the inflation adjustment to principal is taxed as it accrues, before the investor receives the cash. Even the most carefully engineered "risk-free" instrument still carries at least one dimension of risk once you look closely enough. This is worth internalizing as a general principle rather than a footnote about one specific security: every instrument marketed or modeled as risk-free is risk-free along some dimensions and not others, and the useful question is never "is this safe," but "safe from which specific risk, over which specific horizon, for which specific investor." Asked that way, the question stops sounding rhetorical and starts producing a genuinely different answer depending on who is asking it and what they actually need the money for. A retiree drawing down a portfolio for living expenses, a young saver decades from needing the money, and a business owner holding working capital for payroll can each reach a different, individually correct conclusion about the same underlying security, which is exactly the point, and no single product, however it is marketed, can be safe for all three of them simultaneously.
Choosing your own risk-free asset
In a real portfolio, the practical task is choosing the closest available risk-free proxy for each specific time horizon you are managing money against, rather than picking one "safe" bucket for everything. Money needed within the next one to two years, an emergency fund, a planned tax payment, a home down payment, belongs in genuinely short-duration instruments: Treasury bills, a high-yield savings account, or a money market fund holding short paper. Money earmarked for a goal 10 or more years out can tolerate a longer-duration or inflation-linked instrument as its "safe" anchor, since the investor has time to ride out interim price swings.
Bank deposits deserve a specific caution here. FDIC insurance covers up to $250,000 per depositor, per bank, per ownership category, and balances above that limit at a single institution are not government-backed at all, they carry the credit risk of the bank itself, however unlikely default may seem for a large institution in ordinary times. High-net-worth households, and small business owners keeping working capital at a single bank, are the investors most likely to accidentally exceed this threshold without realizing their "safe" cash has quietly become a credit exposure.
High earners saving for a specific near-term milestone, a home down payment, a tax bill on a large bonus or vested equity award, a planned sabbatical, are also the group most likely to have genuinely large sums sitting in a "safe" bucket at any given time, which makes getting the maturity match right worth real attention rather than defaulting to whatever account happens to be convenient. A physician or attorney setting aside $400,000 for a home purchase eighteen months out, for instance, is better served by a mix of Treasury bills and a few insured accounts at different institutions than by either a single bank balance above insurance limits or, worse, a long-duration bond fund exposed to the exact price risk this section describes.
Actionable breakdown
- Match maturity to your actual horizon, not to a generic "safe" label.
- Use bills or a money market fund for goals inside two years.
- Use longer or inflation-linked instruments only for longer horizons.
- Use the short-term Treasury bill yield as your reference risk-free rate.
- Do not substitute a 10-year yield in return calculations without adjusting for its risk.
- Separate default risk from price (duration) risk explicitly.
- A government bond has near-zero default risk but real price risk before maturity.
- Check FDIC and SIPC coverage limits on large cash balances.
- Split deposits across institutions once you approach $250,000 per bank.
Common pitfalls
Investors commonly assume any government-issued bond is risk-free simply because a government stands behind it, ignoring the price volatility a long-duration bond can experience well before maturity if it needs to be sold early. This mistake is especially costly when the money was earmarked for a near-term, non-negotiable goal.
A second pitfall is forgetting that even the safest nominal asset carries inflation risk. A return that is certain in dollar terms is not certain in purchasing power terms, so the standard "risk-free" label refers to nominal, not real, safety, and long-term savers who ignore this distinction can watch a technically riskless account quietly lose ground to rising prices for years.
A third pitfall is treating bank deposits as unconditionally safe regardless of balance size, exposing amounts above insurance limits to a form of credit risk that most depositors never intend to take on. A fourth pitfall, common among investors who feel burned by a past bond price decline, is overcorrecting into money market funds for every dollar regardless of horizon, which trades away a meaningful amount of expected return on long-horizon money purely to avoid a form of price risk that a properly matched, longer-duration instrument would have handled without issue.
The bottom line
Short-term Treasury bills are the standard risk-free asset because their return is certain over their own maturity, but that certainty only transfers to your portfolio when the instrument's maturity actually matches your investment horizon.
Related reading: how bonds work, cash and emergency funds, market history and inflation, the capital allocation line, building a two-asset portfolio.