Weather-Sensitive Demand

What it is

That the weather moves what people buy is an old assumption. Martha Starr-McCluer opened her 2000 Federal Reserve paper by describing the common view that monthly swings in consumer spending are caused by weather. The measurable version of the idea is narrower. It asks how daily or weekly sales at a given store change when temperature, sunshine, rain or snow differ from what is usual for the date, and it asks whether the change is the same size in every store and for every kind of product.

The recent studies say it is not. Florian Badorf and a co-author fitted a random coefficient model, which lets each store have its own response, to daily sales in 673 brick-and-mortar stores. The model allowed non-linear effects and seasonal differences across several weather measures. The effect of weather reached 23.1 per cent depending on location and 40.7 per cent depending on what the paper calls the sales theme, and the direction varied as well as the size. Models that assumed a straight-line response misjudged the effect of extreme good or bad weather.

Online retail shows the same dependence, with a different sign. In a 2017 study led by Sebastian Steinker, sunshine, temperature and rain significantly affected daily orders at a European online fashion retailer, especially in summer, at weekends and in extreme weather. Good weather cut orders strongly and bad weather raised them by much less.

Brigitte Roth Tran's 2022 Federal Reserve Bank of San Francisco letter adds two details from the stores of one apparel and sporting goods brand. Snowfall cut sales at both indoor and outdoor shopping centres. Stores in places with more experience of a kind of weather responded less to it: for precipitation over two inches, the average sales decline was about 80 per cent smaller where experience was above the median.

In effect

The practical use is forecasting. Steinker's team found that adding weather data cut forecast errors by 8.6 to 12.2 per cent on average, and by up to 50.6 per cent on summer weekends, compared with a baseline that ignored it. Badorf and his co-author found that weather forecasts improved daily sales forecasts up to seven days ahead, with the benefit shrinking beyond that. See What a Weather Forecast Does for Sales.

The signal often sits at the level of the product. Linda Dillman, Wal-Mart's chief information officer, told The New York Times in 2004 that strawberry Pop-Tarts sell at about seven times their normal rate ahead of a hurricane, and that beer was the top seller before one. That is an executive's statement with no data shown, and it is used here only to show where retailers look: particular products before a particular event, not the total at the till.

Why weather moves shoppers at all is a separate question. What the Weather Does to Your Spending covers it, and finds the mood explanation weak and the evidence on what people imagine they will need stronger. The effects described here are measured on sales and do not depend on which explanation is right.

What it does not say

It does not give a national figure. The studies measure particular retailers, and the response differs by location and product, so a coefficient estimated on one set of shops should not be applied to another.

It does not say the sign is the same everywhere. Steinker found good weather cut online orders at a European fashion retailer, while Roth Tran reports that online sales of the brand she studied rose on good-weather days. The retailers, countries and years differ, and neither study can say which difference matters.

It does not say the effect is permanent. Roth Tran found good-weather gains were offset by lower sales in the following weeks, which is the subject of demand displacement.

And the evidence is thinner than the confidence of the trade press suggests. Badorf's, Steinker's and Starr-McCluer's papers were read here at the level of their published abstracts. Starr-McCluer's abstract gives no figure for the size of the effect. Dillman's statement is testimony, not a study.

---