Inflation Risk

Inflation risk is the risk that rising prices reduce what an investment's future payments will buy. The SEC names it among the risks every investor carries and states the mechanism plainly: "Inflation reduces purchasing power, which is a risk for investors receiving a fixed rate of interest."

The refinement that makes the idea usable is that the risk is not the inflation itself. It is the part of the inflation nobody had priced in. A bond issued today is bought and sold by people who already have a view about what prices will do over its life, and that view is embedded in the yield they accept. If inflation then runs at exactly the expected rate, the buyer receives precisely the deal they knowingly made, however much purchasing power the nominal dollars lost along the way. What harms the holder is inflation arriving higher than expected, and what benefits them is inflation arriving lower. The material on inflation covers what inflation is, how it is measured and which assets have historically kept pace; this page is about the gap between what was expected and what happened.

Expected inflation is a price, and it can be observed. The gap between the yield on a nominal Treasury security and the yield on an inflation-linked one of the same maturity is a market reading of expected inflation over that horizon, and Federal Reserve banks publish estimates of it. The Cleveland Fed goes further and separates the expectation from the compensation for being wrong about it, describing its inflation risk premium as "a measure of the premium investors require for the possibility that inflation may rise or fall more than they expect over the period in which they hold a bond," and as "an assessment of the risk of unexpected changes in inflation." Two distinct things are therefore being paid for in a nominal bond's yield: the inflation people expect, and the fact that they might be wrong.

The surprise does not arrive as slow erosion. It arrives as a repricing. The intuitive picture of inflation risk is a coupon quietly buying less each year, and for a holder who keeps the bond to maturity that is exactly what happens. But for anyone marking the position to market, the loss lands the moment expectations change rather than over the following decade, because the market immediately demands a higher yield to compensate for the new outlook, and a higher yield means a lower price on every bond already outstanding. This is why inflation risk and interest rate risk are usually the same event seen from two angles: the inflation news is the reason rates moved. A bondholder who treats them as separate exposures will be surprised twice by one event.

Direction matters, and the risk runs both ways. Inflation coming in below what was priced hands the bondholder a better real return than they bargained for, and hands the borrower a heavier real debt than they bargained for. That symmetry is why it is a risk rather than merely a cost. A fixed-rate mortgage is the household's most common position on the other side of it, and a household with a long fixed mortgage and a long nominal bond portfolio holds the same bet twice in opposite directions.

The exposure grows with the term, which produces an inversion worth noticing. A 30-day obligation gives inflation almost no time to be surprising, and the money can be reinvested at whatever rates then prevail. A 30-year fixed nominal payment gives it three decades. So the securities with the least credit risk are typically the ones with the most inflation risk, because it is the strongest issuers that borrow for thirty years at a fixed rate. Describing a long Treasury bond as the safe holding is describing one risk and ignoring the one that most threatens a multi-decade plan.

What inflation-linked securities actually do, stated precisely. Their adjustment tracks realized inflation, so they remove the surprise from the equation and leave the holder with a real return agreed at purchase. What they do not do is guarantee a gain when inflation is high, because their prices move with real interest rates: if real rates rise at the same time as inflation, an inflation-linked holding can fall in value while doing exactly what it was designed to do. They also track a broad published index rather than any household's own spending, which is a second, smaller gap between the hedge and the exposure.

Start with the yield on a nominal bond. It contains a real return, the market's expectation of inflation over the bond's life, and compensation for the chance that the expectation is wrong. What the holder eventually earns in real terms depends on which of those turns out to be right.

A hypothetical example with invented figures. Anneke buys a 10-year bond yielding 4.5 percent at a time when the market expects inflation to average 2.5 percent over the decade.

If inflation averages 2.5 percent, her real return is (1.045 ÷ 1.025) − 1 = about 1.95 percent a year. She got what she signed up for. The nominal dollars bought less each year, and none of that was a risk that materialized, because it was in the price when she bought.

If inflation averages 4.5 percent instead, her real return is (1.045 ÷ 1.045) − 1 = 0 percent. Ten years of lending produced no gain in purchasing power at all.

If inflation averages 1.0 percent, her real return is (1.045 ÷ 1.01) − 1 = about 3.47 percent, considerably better than she expected, and the borrower's real cost rose by the same amount.

Now the timing. Suppose that in year two the market's expectation for the remaining eight years jumps from 2.5 percent to 4.5 percent, so buyers now demand roughly 2 percentage points more yield. If Anneke's bond has a modified duration of about 7 at that point, its price falls by roughly 7 × 2 = 14 percent. Her purchasing power has not yet been eroded by a single month of actual price rises. The market simply revalued the promise for the inflation it now expects, all at once, on the day the expectation changed. Holding the bond to maturity avoids the printed loss and delivers the real return in the second line above, which is the same news arriving slowly.

The quick way to approximate a real return, subtracting inflation from the nominal rate, gets close enough for a rough read and is not exact. The division above is the correct method.

Pros (of framing it as expected versus unexpected)

• It explains why a bond yielding more than the inflation rate is not automatically a good deal, since the extra was already the point.
• It says what inflation-linked securities are actually buying, which is protection against being wrong rather than protection against inflation as such.
• It shows why inflation risk and interest rate risk arrive together, which stops a bondholder counting the same event twice.
• It makes the term of a holding, rather than the identity of the issuer, the variable that controls the exposure.

Cons (and limits of the concept)

• Expected inflation is estimated from market prices, and those estimates are themselves uncertain and revised.
• A household's own inflation rate can differ from the published index that every hedge is built around, so even a perfect hedge is approximate.
• The exposure cannot be removed without accepting something else: shorter maturities carry reinvestment risk and inflation-linked securities carry real rate risk.
• It is a poor description of the risk faced by someone with a large fixed-rate debt, who is on the opposite side of it.