Months of Inventory: How the Number Is Built and Where It Breaks

Months of inventory is the metric most often used to declare a market favorable to buyers or to sellers, and it is a ratio rather than a measurement. Ratios deserve a harder look than measurements, because they can move for reasons in the numerator, reasons in the denominator, or reasons in both at once, and the published figure never tells you which.

This page contains no current reading, deliberately. What follows is how the number is constructed and where its construction fails, which is the part that stays true.

The construction

Take the count of homes available for sale at a moment. Divide by the rate at which homes are being absorbed, expressed as a number per month. The result is expressed in months and is usually described as how long it would take to sell every available home if nothing new arrived and the current pace continued.

Read that description again, because it contains two assumptions that are never true: that no new listings appear, and that the absorption pace holds. The figure is a snapshot dressed as a projection. It is best understood not as a forecast of anything, but as a compact way of expressing the current relationship between supply and demand.

Where the numerator gets slippery

"Homes available for sale" sounds unambiguous and is not.

Publishers differ on whether to count listings that are under contract but not yet closed, whether to count homes marked as available for backup offers, whether to include listings in pre-market or coming-soon status, and how to treat withdrawn listings that return. Each choice moves the count, sometimes substantially, and reports rarely explain which convention they used.

The count is also a point-in-time snapshot, usually taken at a month's end. Anything that arrived and departed between snapshots is invisible. In a market where well-prepared homes can move quickly, a snapshot systematically undercounts the supply that genuinely existed during the period, because the fastest-moving inventory never appears in it.

Where the denominator gets slippery

The absorption rate is usually a count of closings over a recent period, which imports the whole lag problem: closings describe agreements reached weeks earlier, so the denominator lags the numerator. A ratio of a current snapshot to a historical pace is comparing two different moments, and the two are being divided as though they were simultaneous. That mismatch is set out in list date, contract date, close date.

Some publishers instead use contracts signed, which is more current and more volatile. Neither choice is wrong, but they produce different figures from identical facts, and comparing across sources without knowing which was used is meaningless.

Why small markets break the ratio

Ratios become unstable when either side is small, and in a low-turnover, built-out city both sides are small.

The arithmetic is unforgiving. When the denominator is a modest count, a change of a few transactions swings the result noticeably. A period in which a handful more or fewer homes happened to close will move the published months-of-inventory figure even though the number of available homes and the underlying appetite of buyers were unchanged.

The result is a gauge that looks precise, because it is quoted to a decimal, while resting on a base thin enough that the decimal is not meaningful. This is the same small-sample fragility that afflicts price statistics, described in what a median home price hides, and it applies with equal force to supply gauges.

The segmentation problem

A citywide figure blends submarkets that behave nothing alike. Claremont holds entry-level and mid-range homes near sought-after school boundaries, older stock in and around the Village, and hillside properties in Claraboya and the upper northern streets. Those tiers have different buyer pools, different typical timelines, and different supply behavior.

A single citywide ratio can therefore sit at a level that describes neither the tier where competition is real nor the tier where homes take longer, while implying a uniform condition that does not exist anywhere in the city. If you are making a decision about a hillside property, a citywide supply figure dominated by activity elsewhere is close to useless. Segment or discard.

What the ratio is genuinely good for

Two uses survive all of the above.

The first is DIRECTION over time within a consistent segment. If the same publisher, using the same conventions on the same segment, shows the ratio moving steadily one way across several periods, that is real information about the balance between supply and absorption, and it is worth knowing.

The second is confirmation. When the supply ratio, the pace at which homes go under contract, and the frequency of price reductions all point the same way, the agreement of three independent gauges is far stronger than any one of them. Contract activity leads, and is covered in pending sales; the supply ratio confirms.

What it cannot do

It cannot tell you what will happen, because its own definition assumes nothing new arrives, which is never the case.

It cannot tell you what your home is worth, or how quickly your particular home will sell, because it describes a population and your home is one member of it with its own condition, location and preparation.

And it cannot support a confident declaration about market character from a single reading, which is exactly what it is most often used for. A one-period figure in a small market is a coin flip in a suit.

Where to go next

The general habit of checking a statistic's construction before its conclusion is in how to read a Claremont market report, and the disclosures that would let you check this one are in the data behind Claremont market reports.

For why supply here is structurally thin in the first place, read the Claremont housing market, explained, and browse the series on the market reports hub. If you need to know how a specific home would fare rather than how the city is doing, that is a property question and deserves a property answer. Anthony Grynchal has been licensed in California since November 2009.