Liquidity Preference and Preferred Habitat

If the only thing that mattered for bond yields were expectations about future short-term rates, the yield curve should be flat on average, tilting up or down only when the market genuinely expects rates to rise or fall. In reality, the curve is upward sloping most of the time, even when there's no strong expectation that rates will rise. Something besides pure rate expectations is pushing longer yields up — and there are two classic, competing stories for what that something is.

Liquidity preference theory

This theory argues that investors generally prefer shorter maturities because they're more liquid and less exposed to price swings from rate changes — a long bond's price moves much more for the same change in yield than a short bond's does. To persuade investors to hold longer bonds anyway, issuers must offer a term premium, extra yield on top of what expected future rates alone would justify:

$$y_n = \frac{1}{n}E\left[\sum_{t=0}^{n-1} r_t\right] + \text{TP}_n$$

In words: the long-term yield equals the average short rate expected over the life of the bond, plus a term premium $\text{TP}_n$ that grows with maturity $n$ — this premium is why the curve tends to slope upward even when short rates aren't expected to move.

Preferred habitat theory

Preferred habitat agrees the curve isn't purely expectations-driven, but locates the reason differently: it's not that every investor dislikes long maturities — some, like pension funds and insurers with long-dated liabilities, actively prefer them. Instead, each investor class has a natural maturity "habitat" it sticks to for balance-sheet or regulatory reasons, and will only venture outside that habitat if the yield on offer elsewhere is attractive enough to compensate for the mismatch. Supply and demand within each segment — not one universal risk aversion to duration — is what sets relative yields.

Worked example

Suppose pension funds have strong structural demand for 30-year bonds to match long-dated liabilities, pushing 30-year yields down relative to what pure expectations would predict, while a wave of new 10-year corporate issuance floods that segment and pushes 10-year yields up.

1. Pure expectations prediction: if short rates are expected to be flat, the curve should be roughly flat across 10 and 30 years.
2. Preferred habitat effect: heavy pension demand at 30 years pulls that yield down; heavy new supply at 10 years pushes that yield up — producing a curve that's flatter, or even inverted, between 10 and 30 years despite unchanged rate expectations.
3. Practical read: a curve-shape anomaly like this is a segmentation signal, not necessarily a market forecast about future short rates — the two theories give genuinely different trading implications for the same observed curve shape.

What this means in practice

Both theories coexist in how practitioners actually think about the curve: liquidity preference explains the average upward slope and the general existence of a term premium, while preferred habitat explains specific kinks and hump shapes that pure expectations or a uniform term premium can't account for — like why the belly of the curve richens ahead of pension-driven long-end demand, or why heavy Treasury issuance at a specific maturity cheapens that point relative to its neighbors.