Research / Festival & Ceremonial Spending
The Hypothesis
The importance of weddings and family events shows up in the desire to spend on them, as much as in the spending itself.
The third of our founding questions, and the one most easily mismeasured. Count only rupees spent and you conclude that poorer households care less about weddings. Ask them, and you find the opposite.
Festival and wedding spending is the part of Indian household finance most often written off as irrational. A family without savings borrows for a function. An economist calls it waste. That framing assumes the spending is the whole story.
We wanted to separate two things that look identical in a budget: what a household actually spends, and what it wishes it could. If desire simply tracked means, households without money would never describe wanting a grand event at all. So we asked them directly.
Three frameworks explain why visible, socially significant spending carries meaning beyond the occasion.
Veblen's 1899 argument that some spending exists to be seen. The cost isn't a side effect; it's the signal. A function everyone can tell was expensive does work that a cheaper one cannot. (Veblen, 1899)
Spence (1973) formalised why costly signals are credible: they're expensive precisely so they can't be faked. A wedding is one of the few moments a household's standing becomes publicly legible to an entire network at once. (Spence, 1973)
Akerlof and Kranton (2000) added the piece that matters most here: people act to match who they believe they are, not only to impress others. That distinction is exactly what our data turns on. (Akerlof & Kranton, 2000)
The questions were built to catch the gap between wanting and affording. Verbatim examples:
"During festivals or family functions, what do you usually spend more on?"
"When a festival, wedding or family function is coming up, does your household decide a budget in advance, or spend according to what the occasion requires?"
"Suppose your family could organise a simpler wedding or function and save the remaining money. Would everyone be comfortable with that, or would there be concern about what relatives and neighbours might think? Why?"
"In your family or social circle, are there expectations around spending on weddings, festivals, gifts, charity, temple giving, or social events?"
"What is one thing you never hesitate to spend money on for your family?"
Full instruments on the questionnaires page.
"Everyone has a desire [to hold the function grandly]... we'll decide according to whatever budget we have."
Middle-income respondent. The order of that sentence is the finding.
Describes wanting to invite large numbers of guests, well beyond what the household can currently fund.
Tulia Bahadur, middle-income segment
Calls heavy wedding spending a "sunk cost," and chooses a smaller event deliberately.
Naga, high-income segment
Opted for a court marriage, with the means to have done otherwise.
Josef Chellathurey, high-income segment
90 respondents, Chennai, May to June 2026.
Among respondents asked a directly comparable question, 89% confirm that festival and wedding spending genuinely matters to them. Not that they spend heavily. That they want to, whether or not the household can currently act on it.
Across all three hypotheses
Headline confirmation rates, pooled across 90 respondents.
The desire to celebrate well is the most widely shared finding in the entire study, marginally ahead of women's authority over gold.
One middle-income respondent states it almost as a formula: everyone has a desire to hold the function grandly, immediately followed by, we'll decide according to whatever budget we have. The sequence is the result. Desire comes first and is unconditional. The budget only limits how much of it gets acted on.
Respondents who deliberately choose a smaller event describe it as a choice made against a default. Calling a big wedding a sunk cost, or opting for a court marriage, only reads as a decision because the expected thing was to spend generously. Deviation proves the norm.
Formal advance budgeting for these occasions is rare in every segment. Most respondents describe reactive, as-needed spending. That is a caveat about planning mechanics, not about desire: households treat spending well on the occasion as a given, then manage it against whatever constraints exist when it arrives.
Households buying gold years ahead for a daughter's wedding are doing all three things at once: holding a contingency asset, acting on a decision women largely drive, and planning long in advance to celebrate an occasion well. Saving toward a specific future event for years is itself an expression of desire. See the gold study and the purse strings study for the other two angles on the same behaviour.
The conclusion
The desire to host weddings and festivals grandly is present and strongly held across income levels, holding even where households currently lack the means to act on it.
Counting rupees alone would have produced the wrong answer. The clearest thread in the dataset is the gold set aside for a daughter's wedding: contingency planning, a decision women disproportionately drive, and a desire to celebrate the occasion well that households save toward years in advance, all in one behaviour.
The main constraint is size. Three samples of 30, from in and around Chennai across two months, are not statistically representative of South India or of any income group. These findings describe patterns in this dataset.
The timing limitation bites hardest on this hypothesis. Interviews ran across a single two-month window, which does not capture how spending intentions shift across a full festival and wedding season. Asking the same household in October and in January might produce different answers, and we can't rule that out.
Stated desire is also self-reported, and reporting a wish costs nothing. We treat the consistency of the pattern across segments and its appearance in unprompted answers as the evidence, rather than any single respondent's claim.
Following households across an entire wedding and festival calendar is the obvious next step. Subscribe to follow it.
Participate in the study