What the Weather Does to Your Spending
Sunshine is supposed to lift the mood and loosen the wallet. The largest studies find little effect on life satisfaction, and a clearer one on what people imagine they will need.
Key takeaways
- In Norbert Schwarz and Gerald Clore's 1983 study, people called on sunny days rated their lives as more satisfying, but only when nobody had mentioned the weather.
- Across 1,071,290 Americans, Richard Lucas and Nicole Lawless found in 2013 that the weather did not reliably affect judgements of life satisfaction.
- Michael Conlin and colleagues found in 2007 that a 30-degree Fahrenheit colder ordering day raised the probability of returning cold-weather catalogue items by 3.95 per cent.
Does a sunny morning make you spend more freely? Most people would say yes. Richard Shotton's The Choice Factory gives a chapter to mood, arguing that people in a good mood notice more advertising, like it more and examine it less, and advising brands to buy the moments when audiences are likely to be happy. The weather is the mood shifter nobody controls and everybody feels. The research on it is older than the marketing advice, and it divides in a way the advice does not anticipate, between what sunshine does to how we feel and what the sky does to what we buy.
A feeling that forgets where it came from
The starting point is a 1983 study in the Journal of Personality and Social Psychology by Norbert Schwarz and Gerald Clore. An interviewer telephoned students at random on sunny days and on rainy ones, pretended to be calling from out of town, and asked how satisfied they were with their lives. People called on sunny days reported more satisfaction with life as a whole. When the interviewer first asked, as an aside, how the weather was down there, the difference disappeared.
That second result carries the argument. Schwarz and Clore argued that people use their passing mood as information when they judge something large and vague, such as their life. Once the cause of the mood is pointed out, the feeling is set aside as irrelevant. The idea became known as mood as information, and it explains why a bright morning might colour judgements that have nothing to do with the sky, and why it would stop doing so the moment anyone mentioned it.
Shotton's chapter, at pages 57 to 61, applies the same family of ideas to advertising. His own evidence includes a survey in which 21 per cent of respondents who rated their mood at seven or more out of ten liked the ad they were shown, against 13 per cent of those at six or below. The survey asked about the ad first and the mood second, so it cannot say which came first, and it says nothing about weather. His advice follows from it: buy cinema over commuting, weekends over weekdays, and match an ad's energy to its audience's likely mood.
A million interviews against a sunny day
The 1983 study was small. As Richard Lucas and Nicole Lawless later pointed out, it had 84 participants in total, 14 to a group, and its central result depended heavily on one of those groups. Small studies of this kind can produce large effects by chance, and this one became a classic of social psychology before anyone had tried it at scale.
Lucas and Lawless did, in the same journal in 2013. They linked the answers of 1,071,290 Americans to a national health survey run between 2005 and 2009, which included a question on life satisfaction, to the weather in each respondent's metropolitan area on the day they were interviewed. Most weather conditions had no relationship with how satisfied people said they were. The few that did produced effects the authors described as very small, detectable only because the sample was so large. Their conclusion was that the weather does not reliably affect judgements of life satisfaction.
Lucas and Lawless were careful about the limits of that finding. They could not rule out effects altogether, and a telephone survey across a whole country tests something slightly different from a phone call made on a sunny or a rainy day, as the original calls were. Schwarz and Clore's broader account of mood as information has been developed over decades, including in their own 2003 review, and a null result about weather and life satisfaction does not undo the idea that feelings feed judgements.
On the specific question of whether sunshine changes how people rate their lives, the larger study is the better guide. A million interviews make a better guide than 84, and the sunny-day effect, if it exists, is too small to matter for anyone planning a campaign or a budget.
Where sunshine still seems to move money
Money is a different matter, and here the evidence is more mixed. David Hirshleifer and Tyler Shumway looked at morning sunshine in the city of each country's leading stock exchange, across 26 countries from 1982 to 1997, and compared it with that day's market returns. Their 2003 paper in the Journal of Finance found sunshine strongly correlated with returns, while rain and snow added nothing once sunshine was taken into account.
Their own paper set out the practical limit. A trading strategy built on the weather made sense only for a trader with very low costs, because it required frequent trades, and fairly modest costs wiped out the gains. That makes the sunshine effect on markets real enough to measure and too small to exploit, which is a fair description of most of what the weather does to money.
The sunshine effect on markets is real enough to measure and too small to exploit.
A 2010 study by Kyle Murray and colleagues in the Journal of Retailing and Consumer Services is the closest test of shopping itself. Across three studies, including six years of daily sales at a small retailer selling a single category of product, the authors found that more sunlight was linked to more spending. In a laboratory study, people exposed to artificial sunlight were willing to pay more for a product to be used later, and the effect ran through a reduction in bad mood rather than an increase in good mood. That is closer to Shotton's claim than anything else here, though it describes relief from gloom rather than a cheerful audience that has stopped paying attention.
The coat you order in a cold snap
The strongest evidence that weather changes what people buy has little to do with mood at all. It comes from economists studying projection bias, the tendency to assume that how we feel now is how we will feel later. People understand that their tastes change, but they underestimate by how much.
Michael Conlin and colleagues tested this in a 2007 paper in the American Economic Review using catalogue orders of cold-weather items. If the temperature on the day of ordering was 30 degrees Fahrenheit lower, the probability that the item would later be sent back rose by 3.95 per cent. The authors read this as projection bias: the cold on the day of ordering shaped a decision about months of future use more than it should have, and some buyers changed their minds once the item arrived.
Meghan Busse and colleagues found the same pattern in a far larger setting. Their 2015 paper in the Quarterly Journal of Economics used data on more than 40 million vehicle transactions and found that the choice to buy a convertible or a four-wheel drive depended heavily on the weather at the time of purchase, in a way that standard economic theory could not explain. They tested a range of rational explanations, found that none fully accounted for the pattern, and pointed to projection bias and salience, how vivid a feature seems at the moment of choice, as the likely mechanisms.
These two papers suggest that the weather works on spending less through how cheerful we feel and more through what we picture ourselves needing. A cold morning makes the warm coat seem essential. A warm Saturday makes the convertible look like a car for every day. Neither is a mood effect in Shotton's sense.
The weather works on spending less through how cheerful we feel and more through what we picture ourselves needing.
If projection bias is the mechanism, the defence is unglamorous. Leave a day between seeing a seasonal item and buying it, ask whether you would want it on an ordinary day in the middle of next month, and read the returns policy before the cold or the heat makes the decision for you. None of this has been tested as an intervention, but it fits what the field studies found.
That leaves two accounts of the same sunny Saturday. In one, good weather lifts the mood, the mood lowers the guard, and people buy more of whatever they are shown. In the other, the weather changes the picture of next week, and people buy for a future that looks like today. The large studies of mood point away from the first. The field data on purchases point towards the second. Murray's artificial-sunlight study sits awkwardly between them, and separating the two would take a study that measured both mood and expected use on a real shop floor. When the basket fills on a sunny day, which account is at work?
Common questions
Does sunny weather make people spend more?
Some evidence says yes. Kyle Murray and colleagues found in 2010 that more sunlight was linked to more spending, through reduced bad mood. But Richard Lucas and Nicole Lawless found in 2013, across 1,071,290 Americans, that weather did not reliably affect life satisfaction, which weakens the simple good-mood account.
What is projection bias in shopping?
Projection bias is assuming that how you feel now is how you will feel later. Michael Conlin and colleagues found in 2007 that a 30-degree Fahrenheit colder ordering day raised the probability of returning cold-weather catalogue items by 3.95 per cent.
Sources
Schwarz, N., & Clore, G. L. (1983). “Mood, misattribution, and judgments of well-being: informative and directive functions of affective states.” Journal of Personality and Social Psychology, 45(3), 513-523. doi:10.1037/0022-3514.45.3.513. Abstract read (University of Michigan ISR record): more happiness and satisfaction with life in a good mood; the negative impact of bad moods eliminated when participants were induced to attribute their feelings to an external source. Weather procedure read in Schwarz and Clore's own description (Psychological Inquiry, 2003): random student numbers assigned to sunny or rainy days, an interviewer posing as calling from out of town, and the aside 'By the way, how's the weather down there?'; more satisfied on sunny days 'only when their attention was not drawn to the weather'.
Schwarz, N., & Clore, G. L. (2003). “Mood as information: 20 years later.” Psychological Inquiry, 14(3-4), 296-303. doi:10.1207/S15327965PLI1403&4_20. Full text read (author-hosted PDF), for the description of the 1983 procedure.
Lucas, R. E., & Lawless, N. M. (2013). “Does life seem better on a sunny day? Examining the association between daily weather conditions and life satisfaction judgments.” Journal of Personality and Social Psychology, 104(5), 872-884. doi:10.1037/a0032124. Full text read: Behavioral Risk Factor Surveillance System waves 2005 to 2009; 1,071,290 respondents in the primary analyses; most weather conditions unrelated to life satisfaction; significant effects very small; weather does not reliably affect life satisfaction judgements; authors cannot rule out effects entirely. Their description of Schwarz and Clore (1983): total sample 84, 14 per cell in a 2 by 3 design, interaction driven by one cell.
Hirshleifer, D., & Shumway, T. (2003). “Good day sunshine: stock returns and the weather.” Journal of Finance, 58(3), 1009-1032. doi:10.1111/1540-6261.00556. Abstract read (BYU ScholarsArchive): morning sunshine in the city of each country's leading exchange and daily index returns across 26 countries, 1982 to 1997; sunshine strongly significantly correlated with returns; rain and snow unrelated after controlling for sunshine; weather-based strategies optimal only with very low transaction costs, and fairly modest costs eliminate the gains. Affiliations from Ohio State University's news release (read).
Murray, K. B., Di Muro, F., Finn, A., & Popkowski Leszczyc, P. (2010). “The effect of weather on consumer spending.” Journal of Retailing and Consumer Services, 17(6), 512-520. doi:10.1016/j.jretconser.2010.08.006. Abstract read (RePEc): three mixed-methods studies in lab and field; as sunlight exposure increases, negative affect decreases and spending tends to increase. Study detail read in the authors' extended abstract in Advances in Consumer Research, vol. 35: six years of daily sales from one small retailer specialising in a single product category; a laboratory study with artificial sunlight and willingness to pay for a product to be consumed in the future, fully mediated by negative affect.
Conlin, M., O'Donoghue, T., & Vogelsang, T. J. (2007). “Projection bias in catalog orders.” American Economic Review, 97(4), 1217-1249. doi:10.1257/aer.97.4.1217. Abstract read: catalogue orders of cold-weather items; evidence of projection bias over the weather; 'if the order-date temperature declines by 30 degrees F, the return probability increases by 3.95 percent.'
Busse, M. R., Pope, D. G., Pope, J. C., & Silva-Risso, J. (2015). “The psychological effect of weather on car purchases.” Quarterly Journal of Economics, 130(1), 371-414. doi:10.1093/qje/qju033. Abstract read: more than 40 million vehicle transactions; the choice of a convertible or a four-wheel drive highly dependent on weather at the time of purchase, inconsistent with classical utility theory; rational explanations considered and none fully explains the effects; projection bias and salience proposed as mechanisms.
Shotton, R. (2018). The Choice Factory: 25 Behavioural Biases That Influence What We Buy. Harriman House. Read in full by this publication, with a reliability audit. Bias 8, Mood, pp. 57 to 61, including the author's own survey at p. 58 (21 per cent of respondents rating their mood seven or more out of ten liked the ad, against 13 per cent at six or below; ad rated before mood).