> ## 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.
# Veblen Effect
> Veblen Effect is extra demand for a good because it bears a higher price. The 1950 curve may slope up on a stretch. A same-fee card is a separate count.
**Category**: Effects
**Type**: Extra demand for a good because it bears a higher price. The qualities inside the good are a separate part of demand
**Origin**: Harvey Leibenstein, *Quarterly Journal of Economics* 64(2), May 1950, pages 183–207. The price channel is defined on page 189. The display he names is Thorstein Veblen, *The Theory of the Leisure Class* (1899)
**Also known as**: The price channel in conspicuous consumption. A Veblen good is a later label for a good on a stretch where this channel outweighs the ordinary pull of a lower price
**Quick Answer** — The Veblen Effect is extra demand for a good because the good bears a higher price. Harvey Leibenstein named that channel in 1950 and tied it to conspicuous consumption. In his diagram, some stretches of the demand curve can slope up, and an income limit still stops the curve. A later credit-card test held the fee fixed and varied whether the card looked platinum. That gap is a visible mark at one price.
## What is the Veblen Effect?
The Veblen Effect is extra demand for a good because the good bears a higher price.
> By the Veblen effect we refer to the phenomenon of conspicuous consumption; to the extent to which the demand for a consumers' good is increased because it bears a higher rather than a lower price.
That sentence is page 189 of Leibenstein's May 1950 paper. He prints his name as H. Leibenstein and his affiliation as Princeton University. The higher price is the channel. Demand that comes from the qualities inherent in the good is what he calls functional demand, and it sits outside this name.
A dinner can be the same meal at two prices. If the higher price is the part other people are meant to notice, the extra desire for that price is the channel on this page. The cooking is a different question. That picture is an illustration. It is not one of the counts below.
The pages next door measure other misses. [Bandwagon Effect](/effects/bandwagon-effect) on this site is adopting a belief or a behavior because many others already have. Leibenstein's own bandwagon is narrower: demand rises because other people are consuming the same good. [Money Illusion](/effects/money-illusion) is a judgment that treats a nominal amount as if inflation had not changed its buying power. His "real price" on this page is only the money the buyer paid.
### The Veblen Effect in 3 Depths
* **Beginner**: A higher price can be the thing other people are meant to see. Write that seen price before you call the object a better product.
* **Practitioner**: Keep three lines. Did the money paid change? Did the price other people are thought to believe change? Did other people's purchases change? Only the second line is this effect.
* **Advanced**: An upward stretch is allowed, and it is not the whole curve. A high enough price hits an income limit. The snob channel alone, in his proof, cannot make the slope positive.
## Origin
The phrase Leibenstein attaches to the name is conspicuous consumption. The book that made that phrase famous, on his account, is Thorstein Veblen, *The Theory of the Leisure Class* (1899). The text read here is Project Gutenberg eBook 833. It is not a check of a Macmillan page number.
In Chapter Three, Conspicuous Leisure, Veblen writes that in order to gain and to hold the esteem of men it is not sufficient merely to possess wealth or power. The wealth or power must be put in evidence, for esteem is awarded only on evidence. In Chapter Four, Conspicuous Consumption, he writes that conspicuous consumption of valuable goods is a means of reputability to the gentleman of leisure. The same chapter says the utility of conspicuous leisure and of conspicuous consumption, for reputability, lies in the waste common to both. One waste is time and effort. The other is goods.
Leibenstein, on page 184, says this treatment was made famous by Veblen, and that Veblen was neither the discoverer nor the first to elaborate the theory. He names John Rae, writing before 1834, and he cites John Rae, *The Sociological Theory of Capital* (Macmillan, London, 1905), especially Chapter XIII and Appendix I, pages 218–276. That 1905 book was not re-read. No sentence from Rae is on this page.
The 1950 article then leaves the sociology and keeps the demand curve. Section II splits demand into functional and nonfunctional. Nonfunctional demand from other people, or from the price tag, is three named effects. The bandwagon effect is the extent to which demand rises because others are also consuming the same commodity. The snob effect is the extent to which demand falls because others are also consuming it, or are consuming more of it. The Veblen Effect is the price. He says the snob effect is a function of the consumption of others, and the Veblen Effect is a function of price. He notes that some writers had folded the two into one "snob behaviour," and that his cut is the one he finds useful in the analysis.
On page 203 he splits one price into two. The real price is the money the consumer paid. The conspicuous price is the price other people think the consumer paid, and a footnote says the more accurate version is the price the consumer thinks other people think he paid. In a market where price information is common knowledge, the two prices match. Where some buyers get a bargain or a special discount, they need not match. Quantity demanded is a function of both.
Figure 4 is a diagram, not a data set. A drop from one labeled price to a lower one has a price effect that raises quantity and a Veblen effect that removes buyers who wanted the higher conspicuous price. The letters on that figure are not counts. He says the resulting curve can slope down, slope up, or mix the two. It depends on which effect is larger at that price change. If the curve is not falling throughout, he expects a backward-S shape in most cases: some price is so high that the income constraint buys nothing, and some low price is low enough that the good ceases to have conspicuous value. That low-price zero is his assumption. He calls the opposite assumption unlikely.
The snob section, on page 201, says the snob effect is never greater than the price effect. If it were, some snobs would be buying more at a higher price, which breaks his definition of snob behaviour. A footnote says the snob effect plus the Veblen Effect, together, can exceed the price effect. The conclusion on page 207 then states three comparisons with a demand curve that has none of these external effects. If the bandwagon effect dominates, the curve is more elastic. If the snob effect dominates, the curve is less elastic. If the Veblen Effect dominates, the curve is less elastic, and some portions may be positively inclined. If the Veblen Effect is absent, the curve slopes down even when the snob effect is important.
## Key Points
Four readings keep the price channel, the other-buyer channels, and the later counts apart. A diagram letter is not a take-up rate.
Leibenstein's real price is the money paid. His conspicuous price is the price the buyer thinks others think was paid. A receipt and a story about the receipt can differ when a discount is private. [Money Illusion](/effects/money-illusion) uses "real" for buying power after inflation. That is a different pair of numbers. A training note that treats his real price as an inflation-adjusted price has changed the 1950 definition.
Demand that rises because others consume the same good is his bandwagon effect. Demand that falls because others consume it is his snob effect. The site page [Bandwagon Effect](/effects/bandwagon-effect) is the broader habit of joining what already looks popular. Neither of those two is a higher price. A full restaurant can be his bandwagon channel. It is not, by itself, this page.
Page 207 allows some portions of the curve to slope up when this effect dominates. The same section says that, without this effect, the curve slopes down no matter how strong the snob effect is. The backward-S sketch adds an income constraint at the top and a low price at which conspicuous value drops out. A slogan that says demand always rises with price has dropped both ends of his sketch.
A test can show that people pay for a mark others will see, while the fee stays the same. That result is about the mark. It becomes this effect only when the higher price itself is what raises demand. The card study below is the first kind of test. [Compromise Effect](/effects/compromise-effect) is a shift toward the middle of three options. A higher price on one good is not that middle.
## Applications
Write the money paid, the price others are thought to believe, and whether anyone else's purchases changed. Then read the percent again.
Before a price increase is credited with extra desire, write the money the buyer pays and the price other people are thought to believe. Leibenstein's page 203 is that split. A private discount can lower the first and leave the second in place. The two are one price only when price information is common knowledge.
If the brief says people want the good because others already have it, write his bandwagon line or his snob line. The first raises demand as others consume more. The second lowers it. The price line on this page stays empty until the price itself is the thing that changed demand.
When two offers carry the same fee and the same services, a higher take-up for the visible version is demand for the mark. It is not a slope against price. The Indonesian upgrade below charged Rp 360,000 on top of the current annual fee in every arm. That number stays on that offer.
Use the 1950 sketch as a warning about range. Some portion may slope up. A high enough price buys nothing, because of the income constraint. A claim that any luxury price can be raised again, with demand sure to follow, has left his diagram.
## Case Study
Leonardo Bursztyn, Bruno Ferman, Stefano Fiorin, Martin Kanz, and Gautam Rao studied a status mark on a credit card. The text read here is "Status Goods: Experimental Evidence from Platinum Credit Cards," *Quarterly Journal of Economics* 133(3), 2018, pages 1561–1595. They work with one Indonesian bank that had about 200,000 credit card customers and three tiers: classic, gold, and platinum. The platinum tier had the highest income rule. The paper says only 10 percent of active customers qualified for it. The experiment's script also told every customer that only 10 percent normally qualify for these services, so beliefs about exclusivity would match.
The bank identified 1,260 clients. Callers reached 835, and those 835 are the sample. Each client heard the offer once. The design holds the services fixed and changes the card people see. In the benefits arm, the customer keeps a gold card that looks like the one already owned, and the platinum services are added to it. In the platinum arm, the customer receives the dark platinum card, and the script says everybody will notice the difference. A third arm uses the platinum card and replaces "randomly chosen" with "as one of our top customers." The authors later call that merit line too weak to move self-image. The add-on price is the same in the arms they compare: Rp 360,000 (their US\$26) on top of the current annual fee. A customer already paying the gold fee of Rp 240,000 would then pay Rp 600,000 in total, the platinum annual fee. A customer whose fee had been waived would start paying Rp 360,000. The paper's footnote says annual fees are often waived for new customers. These offers went to existing gold-card customers.
At that price, take-up is 13.7 percent for the benefits upgrade (N = 271), 21.0 percent for the platinum upgrade (N = 281), and 23.0 percent for the merit line (N = 283). The permutation test of benefits against platinum is p = .025. Merit against platinum is p = .549. Pooling the two platinum arms, take-up is 22.0 percent against 13.7 percent. The regression coefficient on the pooled platinum offer is 0.083, standard error 0.027, in a sample of 835. The permutation p-value is .004 without covariates and .006 with them. The 7.3 point gap is 21.0 against 13.7. The 8.3 point coefficient is the pooled contrast. They are two readings of the same experiment.
The price cut was not randomized. The bank called back people in the benefits arm who had declined after hearing the price, and offered Rp 90,000 off (their US$6). The paper calls that a 25 percent discount. Rp 90,000 is a quarter of the Rp 360,000 add-on. Of that arm, 13.7 percent had accepted, 48.7 percent had declined before hearing the price, and 37.6 percent had declined after hearing it. The bank reached 70 percent of the last group, and 9.9 percent of those reached accepted. Under the paper's assumptions, including that the unreached 30 percent match the 9.9 percent, predicted take-up for the full benefits arm at the discount is 13.7 percent plus 9.9 percent of the 37.6 percent, which prints as 17.4 percent. The discount adds 3.7 points. A two-sided bootstrapped test that the platinum gap equals this discount effect has p = .12. The authors say a second call can overstate or understate the price response. Their calibration, which this page did not re-read in the online appendix, puts the average value of the status aspect at Rp 218,000 a year (their US$15.50), and they call that figure merely suggestive.
A separate survey says 48 percent of platinum customers own another card that offers cash back. The platinum card offers none at restaurants. The authors read the restaurant use of the platinum card as forgoing that cash back. The 48 percent is ownership. It is not a share of dinners. A consumption survey of 362 customers from the transaction file finds only a small increase in restaurant meals, and that increase is not statistically significant. The paper does not print the size. Platinum customers were not shown to go out more. They were shown, in the offer test, to accept a visible card more often at the same fee.
The lesson is a same-fee gap: 21.0 percent against 13.7 percent, p = .025, for a card others are told they will notice. The ordinary price response in this file, with the caveats above, runs toward the discount. It does not show that a higher price raised take-up.
## Boundaries and Failure Modes
The Veblen Effect is the wrong summary in three retellings. The diagram is treated as a measured slope. The same-fee card is treated as that slope. A visibility correlation is treated as a price.
The first boundary is the 1950 sketch. Positively inclined portions are allowed when this effect outweighs the ordinary price effect. They are not required. The snob effect alone cannot produce them, on his page 201. The letters in Figure 4 are not people. [Anchoring Effect](/effects/anchoring-effect) is a number that pulls a later judgment. A price that is wanted because others can see it is not that pull.
The second boundary is the card. The 21.0 percent and the 13.7 percent hold the fee fixed. The discount prediction of 17.4 percent was not a randomized price, and the equality test is p = .12. A later offer, in March 2016, changed who else could hold the platinum card. The bank had cut the platinum income rule from Rp 500 million (their US$35,500) to Rp 300 million (their US$21,300). Callers reached 93 of 180 platinum customers who already had income of at least Rp 500 million. The diamond card was not yet launched. It had the same services and a different color. The fee printed for it is 650,000 (their US\$46), which the paper says is 50,000 more than the platinum fee, plus Rp 10,000 to receive a formal offer. Everyone who said they would sign up agreed to that Rp 10,000. Take-up moves from 21.6 percent to 40.5 percent when customers are told that lower-income customers can now get the platinum card and cannot get the diamond card. The permutation p-value is .068. The introduction rounds the two rates to 22 percent and 41 percent. This page keeps 21.6 and 40.5. That contrast is about who else qualifies. It is his snob neighborhood, a positional externality in the authors' words, and it is not significant at 5 percent.
The third boundary is visibility without a price slope. Ori Heffetz, "A Test of Conspicuous Consumption: Visibility and Income Elasticities," *The Review of Economics and Statistics* 93(4), November 2011, pages 1101–1117, asks 480 U.S. telephone respondents, from May 2004 to February 2005, how quickly they would notice above-average spending on a category. The question is repeated for 31 categories. The elasticity comparison uses 29. Weighted regressions of those total-expenditure elasticities on visibility give R² = 0.18 for the whole population. For the top three income quintiles the R² values run from 0.19 to 0.32. At the bottom two quintiles he hardly finds this pattern. The abstract's "up to one-third" is that upper end. The paper uses income elasticity and total-expenditure elasticity as interchangeable names because its model has no saving. The estimates are total expenditures. None of these R² values is a demand slope against price. [Endowment Effect](/effects/endowment-effect) is a higher value on a good once a person owns it. Ownership is not the conspicuous price.
The export fails in the same place. The 21.0 percent, the 13.7 percent, the 17.4 percent, the Rp 218,000, and the R² of 0.18 are not a handbag, a bottle, a restaurant check, or another year. A team that raises every luxury price, and expects demand to follow, has not read the income constraint. A team that cites the platinum card as proof that a higher price raised demand has not read the fee.
## Common Misconceptions
Three retellings travel with the name. One gives Veblen a demand curve he did not estimate. One folds the snob channel into the price. One reads the card gap as an upward slope.
The 1899 book is about esteem that requires evidence, and about conspicuous consumption as a means of reputability. It is not a table of prices and quantities. The two-word name, and the claim that demand can rise because the price is higher, are Leibenstein's 1950 article. His Figure 4 uses letters, not a sample. He also says Veblen was not the first to elaborate the idea, and he points to Rae. This page did not re-read Rae.
On page 189 the snob effect is other people's consumption, and it decreases demand. The Veblen Effect is the price, and it increases demand. Page 201 says the snob effect alone never exceeds the ordinary price effect, so it does not by itself make the curve slope up. The two together can. The diamond result, 21.6 percent to 40.5 percent with p = .068, is about other people gaining access. It is not a higher price raising demand.
Both main arms paid the Rp 360,000 add-on. The 21.0 percent against 13.7 percent is the visible card. The later Rp 90,000 discount applies to the benefits arm, was not randomized, and the predicted take-up is 17.4 percent. The test against the platinum gap is p = .12. [Decoy Effect](/effects/decoy-effect) is a third option that shifts choice between two others. One card at one fee is not that third option.
## Related Concepts
The Veblen Effect sits beside a crowd, beside a sticky number, beside a third option, beside a nominal amount, and beside a bid that won because it was the most optimistic.
Joining a belief or a behavior because many others already have. Leibenstein's bandwagon is demand that rises because others consume the same good. The channel here is the price.
A number that pulls a later judgment. A price that is wanted because other people can see it is a different object.
A dominated option that moves choice between two others. The card test compares one fee on two faces of the same services.
A nominal sum read as if inflation had not changed what it buys. Leibenstein's real price is the money paid, not that inflation adjustment.
A winning bid that is high because the winner held the most optimistic signal. The display here is the price others are thought to see.
## One-Line Takeaway
Before you raise a price to make a good more wanted, write the price others are thought to see, and check whether the count in front of you held that price fixed.
This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.