> ## Documentation Index
> Fetch the complete documentation index at: https://meta.niceshare.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Lipstick Effect

> Lipstick Effect is a claimed shift toward cosmetics when larger purchases fall. A 2001 sales note and a 2012 lab do not make it a recession gauge.

<Info>
  **Category**: Effects<br />
  **Type**: A claimed shift, in a weaker economy, toward small goods that change appearance, especially cosmetics. A cheap treat that does not change appearance is a separate claim<br />
  **Origin**: The journalists' name, as Hill and coauthors attach it to Nelson (2001). The sales gauge in the account printed by the South Florida Sun-Sentinel on 2 December 2001 is Leonard Lauder's Leading Lipstick Index. The counted lab is Sarah E. Hill, Christopher D. Rodeheffer, Vladas Griskevicius, Kristina Durante, and Andrew Edward White, *Journal of Personality and Social Psychology* 103(2), August 2012, pages 275–291<br />
  **Also known as**: Lipstick index. That name is the sales gauge. It is not a second title for the desire result in the 2012 lab
</Info>

<Note>
  **Quick Answer** — The Lipstick Effect is the claim that, when the economy weakens, desire or spending moves toward small appearance goods, especially cosmetics, while many larger purchases fall. Leonard Lauder, then chairman of Estée Lauder, described a lipstick index in 2001. A 2012 lab found a rise in women's desire for goods judged useful for attracting a mate, and not for coffee. The index is not a recession gauge.
</Note>

## What is the Lipstick Effect?

The Lipstick Effect is a claimed shift, in a weaker economy, toward small goods that change how a person looks, especially cosmetics, while desire or spending for many larger goods falls.

> When lipstick sales go up, people don't want to buy dresses.

That sentence is Leonard Lauder, chairman of Estée Lauder, in the account printed by the South Florida Sun-Sentinel on 2 December 2001. Lipstick, in that piece, sells for as little as $1.99 in a supermarket and for $20 or more in a department store. The popular telling then adds a movie ticket, a coffee, or any other cheap non-essential, because a house or a car is out of reach and the person still wants a comfort. Those extra goods are not the result this page counts.

A dinner out and a tube of lipstick can both be small. Only the second changes appearance. [Mental Accounting](/effects/mental-accounting) is a label on money by source or by use. The comfort label and the appearance label can sit on the same receipt. [Veblen Effect](/effects/veblen-effect) is extra demand because a good bears a higher price. A lower bill is not that channel.

### The Lipstick Effect in 3 Depths

* **Beginner**: In a tight month, one small visible good can still be bought. Write the good before you treat every small purchase as the same move.
* **Practitioner**: Keep three lines. Did a larger bill fall? Did a cosmetic, or another appearance good, rise? Did a cheap non-appearance treat rise too? Only the second line is the shift this page counts.
* **Advanced**: A lab rise in rated desire is not a national sales gauge. A diary rise among younger women is not proof of one motive. The index, the lab, and the diary answer different questions.

## Origin

The name on this page is the journalists' phrase. Hill, Rodeheffer, Griskevicius, Durante, and White, writing in 2012, say journalists dubbed the pattern the lipstick effect, and they cite Nelson (2001). The copy read for the 2001 account is the South Florida Sun-Sentinel of 2 December 2001. That text does not print a byline. It does print Lauder's sentence, and it prints the name of his gauge.

Lauder's Leading Lipstick Index tracks lipstick sales across Estée Lauder brands. The piece says those brands account for about half of prestige cosmetics sales in the United States, and it names Stila, Origins, Bobbi Brown, MAC, and Prescriptives. Since the 11 September 2001 attacks, Lauder says the index is up broadly. He also says it climbed in the 1990 recession. Neither remark is a percent. MAC factories started running extra shifts to produce more lipstick after 11 September. In the three weeks before the piece, MAC lipstick and lip gloss sales at stores open at least a year were 12 percent above the year earlier. Lipstick sales at mass retailers tracked by Information Resources Inc. rose 11 percent from August through October, compared with a year earlier. Borghese lipstick sales were also up 12 percent since mid-September. The same piece says Estée Lauder cut its overall sales expectations, because other business, especially duty-free airport shops, was hurting. The 11 percent and the two 12 percent lines are not one sample.

*The Economist*, 23 January 2009, in the piece "Lip service," calls the lipstick index a term Lauder coined in the 2001 recession, and it treats an 11 percent rise in American lipstick sales that autumn as the story. This page does not read that 11 percent as a second measurement beside the Information Resources window. The same piece says believers trace a 25 percent rise in cosmetic sales to the Depression, and that reliable historical sales figures are hard to find. Data from Kline & Company, in that piece, show lipstick sales sometimes rising in distress and also rising in prosperity. The piece's conclusion is that there is no correlation. Hill and coauthors cite Nancy F. Koehn (2001) for an unexpected cosmetics boom in the Great Depression. That case was not re-read. No Depression percent from it is on this page.

The 2012 paper is the first experimental demonstration they claim. It was published online on 28 May 2012. The issue is August 2012, volume 103, number 2, pages 275–291, DOI 10.1037/a0028657. Hill and Rodeheffer are at Texas Christian University. Griskevicius is at the Carlson School of Management, University of Minnesota. Durante is at the University of Texas at San Antonio. White is at Arizona State University. They use life-history theory and mate-preference research to predict that recession cues raise women's effort to attract mates who have resources, and that the effort shows up as desire for goods that improve appearance. Study 1 is not that experiment. It is 232 months of U.S. unemployment, from the Bureau of Labor Statistics, set beside monthly retail sales from January 1992 to April 2011, from the U.S. Census Bureau.

Relative spending is the category's dollars divided by retail spending across the categories in that month. Unemployment correlates negatively with the furniture share, r(232) = −.79, with electronics, r(232) = −.52, and with leisure and hobby products, r(232) = −.21, each p \< .001. It correlates positively with personal care and cosmetics, r(232) = .52, and with clothing and accessories, r(232) = .20, each p \< .001. The footnote on raw dollars is weaker. Personal care and cosmetics, r(232) = .30, p \< .001, and clothing and accessories, r(232) = .14, p = .04, still rise with unemployment. Furniture falls, r(232) = −.27, p \< .001. Electronics, r(232) = −.03, p = .71, and leisure and hobby products, r(232) = −.03, p = .72, do not. The three non-appearance categories together are r(232) = −.12, p = .076. A share can rise while the raw dollar line barely moves. [Ratchet Effect](/effects/ratchet-effect) on this site is Duesenberry's claim that consumption does not fall all the way back when income does. That claim is not a cosmetics share.

## Key Points

Four readings keep the tube, the cheap treat, the rated desire, and the diary apart. A company index is not a lab mean.

<Steps>
  <Step title="Name the good before the comfort story">
    The 2001 account is about lipstick set against a dress. A popular retelling adds movies, coffee, and any cheap non-essential. Hill and coauthors say journalists often read the pattern as a cheap indulgence. Their Study 4 put coffee in the cheap non-appearance set. Desire for that set did not rise. A comfort story that cannot say which good moved has left the page.
  </Step>

  <Step title="Read a share and a raw dollar line apart">
    In Study 1, the personal-care share moves with unemployment at r(232) = .52. The raw personal-care line is r(232) = .30. Electronics are −.52 as a share and −.03 as raw dollars, and the raw test is p = .71. A chart of shares is not a chart of dollars. [Loss Aversion](/effects/loss-aversion) is a greater weight on a loss than on a matched gain. A smaller share for furniture is not that weight.
  </Step>

  <Step title="Ask whether the good is judged to change appearance">
    In the lab, recession cues raised women's desire for goods tied to appearance and lowered desire for goods that were not. Men in Study 2 wanted less of either set. Study 5 used the same kinds of goods and changed the slogan. The rise sat on the mate-attraction slogan. The other slogan did not move. A logo, by itself, is not the result.
  </Step>

  <Step title="Keep the motive claims from closing the diary">
    The 2012 authors read the lab rise as effort to attract a mate with resources. Their own direct path, after that preference, is still p = .07. A later diary of the Great Recession finds a cosmetics rise among women ages 18 to 40 that is not confined to unmarried women or to employed women. The lab and the diary are not one mechanism. [Diderot Effect](/effects/diderot-effect) is one new object that makes neighboring objects look too plain, so more goods get replaced. A shift from a clothing bill toward a smaller cosmetic is the other direction.
  </Step>
</Steps>

## Applications

Write the good, the bill it might be replacing, and whose sales moved. Then decide whether the note in front of you is a gauge, a rating, or a diary.

<CardGroup cols={2}>
  <Card title="Split the appearance good from the treat" icon="receipt">
    Before a tight month is called this effect, write the good. A cosmetic, a piece of clothing, and a coffee are three lines. The 2001 piece is the first against a dress. Study 4 is the test of the third. A movie ticket has no count on this page.
  </Card>

  <Card title="Keep one company's index inside that company" icon="chart-line">
    Lauder's index is lipstick across Estée Lauder brands, and he said it was up broadly, with no percent. The 11 percent is Information Resources, mass retail, August through October 2001, against the year earlier. The MAC 12 percent is three weeks at stores open at least a year. Do not add those lines into one national rate.
  </Card>

  <Card title="Read a 7-point wish as a wish" icon="scale-unbalanced">
    Study 2 asked students how much they agreed that they would like to buy the item, from 1 to 7. The means below are that rating. They are not receipts, and they are not a shop's week. A buyer who treats 6.19 as a sales jump has changed the scale.
  </Card>

  <Card title="Ask which bill fell before you call it comfort" icon="shirt">
    In the diary draft below, younger women's cosmetics spending rose while the authors point to women's clothing as the bill that gave way, even though cosmetics became dearer relative to that clothing. A household can use that as a question. It is not a rule for the next grocery trip.
  </Card>
</CardGroup>

## Case Study

The counted case is Study 2 of Hill, Rodeheffer, Griskevicius, Durante, and White (2012). One hundred fifty-four university students took part, 82 women and 72 men, ages 18 to 28 (mean 19.47, SD 1.64). Seventy-nine read a recession article and 75 read a control article. The recession text was a modified *Wall Street Journal* article of 18 September 2008, formatted to look like the *New York Times*, 622 words, describing unemployment and scarcity with no end in sight. The control was an architecture article based on the *Wall Street Journal* of 4 December 2009, 557 words. Each person then rated six products on the sentence "I would like to purchase this item," from 1 (strongly disagree) to 7 (strongly agree). Women saw form-fitting jeans, a form-fitting black dress, and lipstick. Men saw form-fitting jeans, a form-fitting polo shirt, and a men's facial cream. Everyone saw a wireless mouse, a stapler, and headphones. A separate group of 44 undergraduates had rated the beauty set as much more likely to be used to attract a mate. Those prerating means are not the purchase means below.

For women, recession versus control interacted with product type, F(1, 80) = 33.50, p \< .001. Desire for the appearance set was 6.19 (SD 0.97) after the recession article and 4.97 (SD 1.60) after the control, F(1, 80) = 17.76, p \< .001, d = 0.92. Desire for the mouse, stapler, and headphones was 3.47 (SD 0.92) against 3.98 (SD 0.96), F(1, 80) = 5.95, p = .02, d = 0.54. The appearance line rose. The other line fell. For men, product type did not interact with the article (ps > .39). Men wanted less of either set after the recession article, 4.07 (SD 1.24) against 4.51 (SD 1.17), F(1, 70) = 3.91, p = .05, d = 0.36. The three-way interaction of sex, article, and product type was F(1, 150) = 9.56, p = .002.

The lesson is a rated wish among these students. After a bleak article, the women wanted the appearance goods more and the other goods less. The men wanted less of both. The authors say this is the first experimental evidence of the lipstick effect, and that the spending pattern in Study 1 is likely to be driven by women. The 6.19 is not a receipt, and it is not the 11 percent from 2001.

## Boundaries and Failure Modes

The Lipstick Effect is the wrong summary in three retellings. Every cheap comfort is treated as the same rise. The dearest beauty good is treated as the one people skip. A sales bump is treated as a date for the recession, or as proof of one motive.

The first boundary is the cheap treat. Study 4 used 64 unmarried women students, 31 in the recession slideshow and 33 in the control, ages 18 to 20 (mean 18.34, SD 0.54). Luxury appearance goods were Coach perfume, a Nordstrom dress, and Seven jeans. Discount versions were Celine Dion perfume, a WalMart dress, and Faded Glory jeans. Cheap goods that do not change appearance were Starbucks coffee, a SIGG travel mug, and colored BIC pens. Luxury desire was 5.86 (SD 0.86) against 5.18 (SD 1.08), F(1, 62) = 7.70, p = .007, d = 0.70. The cheap non-appearance set was 2.86 (SD 1.10) against 3.46 (SD 1.41), F(1, 62) = 3.57, p = .06, d = 0.47. That row is not significant at 5 percent, and the recession mean is the lower one. Discount beauty was 2.77 (SD 0.81) against 3.06 (SD 0.91), F(1, 62) = 1.74, p = .019, d = 0.37. The authors' sentence is that recessions did not increase desire for those discount goods. An F of 1.74 on 1 and 62 degrees of freedom is not the usual companion of p = .019. This page does not repair that pair. The recession mean is still the lower one. Own resource access did not predict interest (all ps ≥ .69) and did not interact with the slideshow (all ps ≥ .38). The sample was mostly middle class. Coffee did not carry the rise.

The second boundary is the slogan, and the mood check beside it. Study 3 used 76 unmarried women, 36 and 40, ages 18 to 24 (mean 19.66). A slideshow of unemployment lines, foreclosure signs, and empty offices raised desire for a tank top, fitted jeans, make-up, and perfume, 5.35 (SD 1.07) against 4.75 (SD 1.18), F(1, 74) = 5.30, p = .02, d = 0.53, and lowered desire for an mp3 player, a laptop, dish soap, and laundry detergent, 3.56 (SD 1.25) against 4.31 (SD 1.37), F(1, 74) = 6.28, p = .01, d = .57. Preference for a partner's resources was 5.36 (SD 1.00) against 4.84 (SD 1.36), F(1, 73) = 4.74, p = .04, d = 0.44. The indirect path through that preference was β = .16, SE = .09, 95 percent CI \[.02, .39]. The direct path after the mediator was β = .47, SE = .26, t(72) = 1.79, p = .07. The authors call the mediation full and also call p = .07 marginal. Positive and negative affect did not differ (ps ≥ .26). Study 5 then held the goods and changed the words, with 72 unmarried women, 35 and 37, ages 18 to 22 (mean 18.59). Jeans, perfume, and boots with a mate-attraction slogan scored 4.71 (SD 0.90) against 4.37 (SD 0.97), F(1, 69) = 5.28, p = .03, d = 0.36. The same kinds of goods with a non-mating slogan scored 4.25 (SD 0.93) against 4.37 (SD 0.97), F(1, 69) = 0.12, p = .73, d = 0.13. At one standard deviation above the women's own long-term mating motivation, the recession contrast was β = .45, SE = .19, t(67) = 4.19, p \< .001. At one standard deviation below, p = .41. Ekaterina Netchaeva and McKenzie Rees, *Psychological Science* 27(8), August 2016, pages 1157–1168, argue from the abstract that economic concern also raises a wish to look competent at work, and that women high in that concern elect a professional appearance more often than a romantic one. That PDF was not re-read. No cell from it is on this page.

The third boundary is the diary, and the gauge. Daniel MacDonald and Yasemin Dildar study the Great Recession in the Bureau of Labor Statistics Consumer Expenditure Diary. The text read here is their working draft, as posted by the University of California, Riverside. The journal typeset, *Journal of Behavioral and Experimental Economics* 86, June 2020, article 101527, was not re-read. The published abstract agrees that average cosmetics expenditures rose among women ages 18 to 40, regardless of marital or employment status, and that the authors read a substitution away from women's clothes. In the draft, the National Bureau of Economic Research window is December 2007 to June 2009, and their recession indicator is 2008 through 2010. Comparing the 2005–2007 average with the 2008–2010 average, total cosmetics expenditures in the diary declined 0.2 percent. Among women ages 18 to 40, the same window is a 2.8 percent rise for unmarried women and a 9.8 percent rise for married women. Among households that bought cosmetics, the recession coefficient is 0.094 log points, about 9.4 percent above the pre-recession average of those buyers. The 0.2 percent, the 2.8 percent, the 9.8 percent, and the 9.4 percent are not one number. The recession years go with lower odds of any cosmetics purchase in the two-week diary, odds ratio 0.91. Cosmetics prices rose 2.9 percent from 2007 to 2010, while the all-items index rose 5.2 percent. Real median household income fell about 7.2 percent, from $59,500 to $55,500 in 2017 dollars, on their citation of the Census. The draft says the price index for women's clothing fell, so cosmetics became dearer relative to clothing, while younger women still shifted share toward cosmetics and away from that clothing. Those share points are not printed here. *The Economist* in 2009 already said the Kline series shows no correlation that would make lipstick a gauge. A team that dates a recession from a lipstick chart has not read that sentence. A team that exports 6.19, 11 percent, 9.4 percent, or r = .52 onto a cafe, a cinema, a household, or another year has not read the sample.

## Common Misconceptions

Three retellings travel with the name. One adds every cheap comfort to the tube. One keeps only the bargain beauty item. One treats a sales bump as a calendar, or as a single motive.

<AccordionGroup>
  <Accordion title="Movies, coffee, and lipstick rise together">
    The comfort story is widespread, and Lauder's 2001 sentence is about a dress, not a movie. Study 4 put Starbucks coffee with a travel mug and colored pens. That set was lower after the recession slideshow, and p = .06 is not a rise. No cinema count is on this page. The Depression 25 percent line is what believers tell *The Economist* in 2009, and that piece says the historical figures are hard to find.
  </Accordion>

  <Accordion title="Only the cheapest beauty item rises">
    Study 4's luxury set rose, 5.86 against 5.18. The discount set was lower in the recession cell. The authors say a cheap indulgence is the journalists' theory, and their result points at goods women rated as more effective for attracting a mate, even when those goods cost more. A shelf of only the lowest price has dropped that contrast.
  </Accordion>

  <Accordion title="A bump dates the recession or proves one motive">
    The 2009 Kline reading is no correlation. Lauder's own index has no percent in the 2001 piece, and Estée Lauder's wider business was cut back in that same account. The lab's direct path remains p = .07. The diary rise among women ages 18 to 40 is not confined to unmarried women, and the employment interaction is not significant. The wish, the share, and the diary dollar line stay three counts.
  </Accordion>
</AccordionGroup>

## Related Concepts

The Lipstick Effect sits beside a higher price that is wanted because it can be seen, beside a label on a small purchase, beside spending that does not fall all the way back, beside the extra weight of a loss, and beside one new object that pulls further objects in.

<CardGroup cols={3}>
  <Card title="Veblen Effect" icon="gem" href="/effects/veblen-effect">
    Extra demand because a good bears a higher price. The shift on this page is not that upward price channel.
  </Card>

  <Card title="Mental Accounting" icon="wallet" href="/effects/mental-accounting">
    A label on money by source or by use. A comfort label and an appearance label can share one receipt.
  </Card>

  <Card title="Ratchet Effect" icon="stairs" href="/effects/ratchet-effect">
    Consumption that does not fall all the way back when income does, in Duesenberry's 1949 claim. A cosmetics share is not that ratchet.
  </Card>

  <Card title="Loss Aversion" icon="exclamation-triangle" href="/effects/loss-aversion">
    A greater weight on a loss than on a matched gain. A smaller furniture share is not that weight.
  </Card>

  <Card title="Diderot Effect" icon="basket-shopping" href="/effects/diderot-effect">
    One new object that makes the goods around it look too plain, so more goods get replaced. A smaller cosmetic in place of a clothing bill runs the other way.
  </Card>
</CardGroup>

## One-Line Takeaway

<Tip>Before a lipstick sales bump is treated as a recession signal, write whose sales moved, whether the good changes appearance, and whether the cheap non-appearance treats moved the same way.</Tip>


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.