Arca
Sign inCreate my account
FeaturesSecurityBetaGuidesDocsSign in
LanguageFR

← Guides

Updated 13 min read

Why capitalism needs you to keep buying

If you spend more than you would like to, the first thing worth knowing is that this is probably not a flaw in your character. It is an asymmetry. On one side, there is you, your attention and the end of your month. On the other, an economy whose growth depends on your next purchase — and which spent more than a trillion dollars on advertising last year, a threshold crossed in 2024, to make sure of it.

This guide names the system and describes its machinery as research documents it — not as documentaries tell it, because criticism of consumption carries legends of its own. Then it shows what actually protects you. No moralising: research has also measured that making people feel guilty changes almost nothing, and that past a certain point it produces the opposite effect.

The engine: an economy that needs this not to stop

Capitalism has one simple property: a company lives on its sales, and a company whose customers are durably satisfied — equipped, content, settled — is a company in decline. So demand has to be renewed. Not out of malice: out of competition. The firm that does not revive desire loses to the one that does. This is what the economist John Kenneth Galbraith called, as early as 1958, the “dependence effect”: in an economy of abundance, production no longer answers pre-existing needs — it manufactures the desires it claims to satisfy. Before him, Thorstein Veblen had described conspicuous consumption in 1899 — spending to display rank; after him, Jean Baudrillard observed that we no longer buy objects but signs. Three eras, one finding: in this system, consumption is not a means that stops once the need is met. It is the cruising speed.

This is not merely a thinker’s theory: the archives hold proof of it. In 1924, lightbulb manufacturers meeting as a cartel fixed at one thousand hours the life of a product that reached two thousand, with fines for factories that were too conscientious — the Osram archives record it. In 1932, a pamphlet proposed making “planned obsolescence” compulsory by law, to revive sales. In 1954, the designer Brooks Stevens turned it into a doctrine of desire: to instil the desire to own something “a little newer, a little sooner than is necessary”. France made it a criminal offence in 2015; no conviction has yet been handed down on that ground. The machine was not invented by apps; they only made it smoother.

One caution before going further: there is no conspiracy. Nobody meets to decide that you will overspend this month — the lightbulb cartel is precisely the documented exception. There are millions of actors who each, separately, optimise their own sales. That is more ordinary than a conspiracy, and far more effective: a conspiracy can be broken up.

The machine today: fewer steps, more urgency, less pain

Here is what that need to sell has built, with the figures behind it.

Removing a step makes people spend more. When a large online retailer offered one-click payment, the customers who adopted it spent 28.5% more and bought 43% more often (Unal & Park, Management Science, 2023). Nothing about their needs had changed; one decision step had disappeared, that is all. Instalment payment works on the same spring: studies converge in saying it makes people spend more — and in France, the share of over-indebtedness cases containing buy-now-pay-later or small credit rose from 1% in 2022 to 17% in 2024 (Banque de France).

Urgency is often manufactured. A Princeton team analysed eleven thousand retail sites: 11% used at least one deceptive device — countdown timers that reset when they reach zero, stock messages generated at random (Mathur et al., 2019). Vendors sell these mechanisms off the shelf. The device works on average — two meta-analyses confirm it for scarcity signals — and it is now illegal on European platforms: article 25 of the Digital Services Act bans interfaces that “deceive or manipulate”. The US Federal Trade Commission obtained $245 million from Epic Games and a $2.5 billion settlement from Amazon on this ground.

Paying hurts less and less. Research calls “pain of paying” the small sting that accompanies spending cash, and which fades as payment becomes disembodied. One point of honesty: the old story that cards make you spend “twice as much” comes from a single experiment in 2001, never recovered at that magnitude. The most recent meta-analysis — 392 studies — concludes that the effect is real but small, and declining as contactless becomes the norm (Schomburgk et al., Journal of Retailing, 2024). The direction is not in dispute; the magnitude is.

Why it does not fill the gap

A system that needs you to buy again has an unexpected ally: your own brain. Because the implicit promise of a purchase — that you will feel durably better — rarely holds, and that gap restarts demand.

The brain separates wanting from liking. Dopamine, contrary to its reputation, is not the molecule of pleasure but of anticipation: the reward circuit fires more at the prospect of getting than at the moment of having (Berridge, 2007; Knutson, 2001, 2007). That is why the parcel on its way excites and the parcel opened disappoints — and why we start again. We get used to what we buy, and faster to objects than to experiences.

On the substance, the literature is unusually consistent: attachment to possessions and to the image they project is associated with lower well-being. The effect is modest, but it has been found across more than 250 independent samples (Dittmar et al., 2014). And three longitudinal studies show the path in reverse: when a person’s materialism falls over the years, their well-being rises (Kasser et al., 2014).

“Retail therapy” does work — briefly, and not for the reason people think. It is not owning that soothes, it is choosing, which restores a sense of control (Rick et al., 2014). Its lasting effectiveness has never been documented. Worse: over six years, seeking a remedy for low mood in purchases increases loneliness, which in turn increases that same reflex (Pieters, 2013).

What other people see

Part of spending does not aim at the object: it aims at rank. This was Veblen’s old observation, and it is now the best measured.

The cleanest demonstration comes from a Dutch lottery that draws entire postal codes. When a household wins, the probability that its immediate neighbours — including those who never played — buy a car within six months rises by about seven percentage points (Kuhn et al., American Economic Review, 2011). Field experiments in Indonesia showed the mechanism at full scale: demand for a “platinum” bank card exceeds demand for a strictly identical product without the marker; and raising customers’ self-esteem is enough to lower that demand (Bursztyn et al., 2018). We buy status when our self-image wavers.

Social media industrialises that comparison. A recent meta-analysis — 55 studies, 28,000 participants — finds a clear association between their use and materialism, especially when following consumption content. Honesty requires saying it: this is a correlation, not proof that one causes the other. But the direction of the feed is known — comparison there runs almost always upward, particularly while scrolling passively.

What does not work

You do not leave a system merely by knowing about it — and research has closed off three popular counter-strategies.

Relying on willpower. The theory of a “willpower muscle” that depletes has collapsed: two large-scale replications — 23 laboratories, then 36 — found nothing (Hagger, 2016; Vohs, 2021). If your strategy is to resist aisle by aisle, you are playing the machine at its strongest game.

Guilt. Small overall effect, and past a threshold, anger and rejection of the message (Peng et al., 2023). That is also why this guide contains none.

Information alone. Conventional financial education explains about 0.1% of the variance in behaviour (Fernandes et al., 2014). Knowing that the system pushes you is not enough; structure is what is needed.

What works: three habits

Nobody is going to shrink global advertising from their kitchen. What you can do is take back your own perimeter — and there, research supports three mechanisms: modest, real, and cumulative.

Partition. Dividing a resource into compartments slows its consumption, because every crossing of a partition reopens a conscious decision (Cheema & Soman, 2008). In a field trial, putting the same savings into two envelopes rather than one increased the amounts kept by 72% (Soman & Cheema, 2011). Money in a single pile is spent continuously; partitioned money is spent by choice. That is the principle behind envelope budgeting.

Decide in advance. Committing before temptation beats resisting in the moment: this is the principle of commitment devices, documented from Philippine savings accounts to the “Save More Tomorrow” plan, which took savings rates from 3.5% to 13.6% in forty months (Thaler & Benartzi, 2004). Giving every euro a role before the month starts belongs to the same family: when the urge arrives, the decision has already been made — and overspending an envelope becomes an event you look at squarely, not a drift you discover too late.

Watch yourself — actively. Self-monitoring is one of the best established effects in behaviour change (22 studies on eating; the first studies on money point the same way — spending reduced when tracking comes with a goal and feedback, French et al., 2020; Huebner et al., 2020). The detail that matters: passive tracking does not protect — in South Korea, plain transaction confirmation texts increased spending among lighter spenders (Kim et al., 2024). Watching money go past is not enough; it has to be watched against a goal — and against the right number, because your bank balance is not your budget.

A temptation budget illustrates all three at once: in the laboratory, setting an allowance reduces indulgences — but only if you can see where you stand (Krishnamurthy & Prokopec, 2010).

What Arca does with this

Arca did not invent these mechanisms; the app equips them, and that is all it promises.

Envelopes are the partition: the month’s money is divided into named parts, and a purchase comes out of a specific envelope — not an undifferentiated pile. Every euro is given a role before it is spent: the decision taken in advance, calmly, rather than in the moment of temptation. The envelope balance is visible at the instant you decide — exactly the condition the laboratory identified for a budget to bite. And spending is entered by hand: a few seconds of deliberate friction, which make each purchase an act you look at rather than a flow you undergo — active tracking, not a passive statement.

And what Arca does not do: no advertising, no notification pushing you to open the app, no counter applying pressure. A budgeting tool that used the devices described above would disqualify itself.

Let us also be as honest as the studies quoted here: no research has measured “envelope budgeting” end to end, and nobody can promise you “X% less spending” — when interventions of this kind move from the laboratory to the real world, their effects shrink. What research establishes is the three habits. Arca makes them practicable, every day, without moralising and without noise. The system will go on wanting you to buy; your envelopes, for their part, know what you decided.

Sources

The engine:

  • Veblen, T. (1899). The Theory of the Leisure Class, ch. IV.
  • Galbraith, J. K. (1958). The Affluent Society, ch. 11 — the dependence effect.
  • Baudrillard, J. (1970). La société de consommation, Denoël (Eng. The Consumer Society, Sage, 1998).
  • Krajewski, M. (2014). “The Great Lightbulb Conspiracy”, IEEE Spectrum — the Phoebus cartel, from the Osram archives.
  • London, B. (1932). Ending the Depression Through Planned Obsolescence.
  • Adamson, G. (2003). Industrial Strength Design: How Brooks Stevens Shaped Your World, MIT Press.
  • French law no. 2015-992 — planned obsolescence as a criminal offence (art. L. 441-2, Consumer Code).
  • WARC (2024) — global advertising spend passing $1 trillion for the first time.

The machine:

  • Unal, M. & Park, Y.-H. (2023). Management Science — one-click payment. doi.org/10.1287/mnsc.2023.4676
  • Banque de France, Observatoire de l’inclusion bancaire — instalment payment in over-indebtedness cases (2022-2024).
  • Di Maggio, M., Katz, J. & Williams, E. (2022). NBER Working Paper 30508 — BNPL raises total spending.
  • Mathur, A. et al. (2019). “Dark Patterns at Scale”, ACM CSCW. doi.org/10.1145/3359183
  • European Commission (2022) — dark patterns, mystery shopping; FTC v. Epic Games (2023) and Amazon (2025); EU Regulation 2022/2065 (DSA), art. 25.
  • Barton, B., Zlatevska, N. & Oppewal, H. (2022). Journal of Retailing — meta-analysis of scarcity signals; converging: Ladeira et al. (2023), Psychology & Marketing.
  • Prelec, D. & Simester, D. (2001). Marketing Letters — the isolated “×2 by card” experiment.
  • Schomburgk, L., Belli, A. & Hoffmann, A. (2024). Journal of Retailing 100(3) — cashless meta-analysis, 392 studies.

Why it does not fill the gap:

What other people see:

  • Kuhn, P. et al. (2011). American Economic Review 101(5) — the postcode lottery.
  • Bursztyn, L. et al. (2018). Quarterly Journal of Economics 133(3) — the “platinum” cards.
  • Liu, Y., Ding, Y. & Chao, M. (2026). Humanities & Social Sciences Communications — meta-analysis, social media and materialism (correlational).
  • Verduyn, P. et al. (2020). Current Opinion in Psychology 36 — upward comparison.

What does not work:

  • Hagger, M. S. et al. (2016) and Vohs, K. D. et al. (2021) — the two large-scale replications of ego depletion.
  • Peng, L. et al. (2023) — meta-analysis of guilt appeals, boomerang effect.
  • Fernandes, D., Lynch, J. G. & Netemeyer, R. G. (2014). Management Science — financial education and the 0.1%.

What works:

  • Cheema, A. & Soman, D. (2008). Journal of Marketing Research 45(6). doi.org/10.1509/jmkr.45.6.665
  • Soman, D. & Cheema, A. (2011). Journal of Marketing Research 48(SPL) — the two-envelope field trial. doi.org/10.1509/jmkr.48.SPL.S14
  • Thaler, R. H. & Benartzi, S. (2004). Journal of Political Economy 112(S1) — “Save More Tomorrow”.
  • Krishnamurthy, P. & Prokopec, S. (2010). Journal of Consumer Research 37(1) — a budget bites only if the balance is visible.
  • Burke, L. E. et al. (2011). JADA 111(1) — self-monitoring, systematic review.
  • French, D., McKillop, D. & Stewart, E. (2020). European Journal of Finance 26(4-5); Huebner, G., Fleisch, E. & Ilic, A. (2020). Computers in Human Behavior 113.
  • Kim, M. et al. (2024). — passive tracking that increased spending. doi.org/10.1177/10949968231189505
  • DellaVigna, S. & Linos, E. (2022). Econometrica 90(1) — effects shrink from laboratory to scale.