Why We Overspend on Gifts

WANG, Sherry Shi | VAN DER LANS, Ralf

Gift-giving is a big business – in the US annual spending on gift-giving reached $159 billion in 2017. But such expenditure may also be symptomatic of overspending. In the UK in 2015, one in 10 people reported experiencing financial difficulties because they spent too much over the Christmas season. Moreover, recipients are often not as impressed as the givers might hope. One estimate shows they assigned between 10 per cent and 33 per cent lower monetary value to gifts than the givers had paid, a phenomenon known as deadweight loss of gift giving.

There is therefore a need to better understand why givers overspend in order to please receivers. Sherry Shi Wang and Ralf van der Lans offer an explanation through a model that captures two factors affecting gift-making decisions: the giver’s inaccurate prediction of the receiver’s preferences, and the price. The authors argue that uncertainties about those preferences can lead givers to have lower price sensitivities and spend more than they might otherwise.

“If uncertainty increases, it becomes more difficult to predict the receiver’s preferences and the giver may therefore decide to put more weight on the signaling value of price,” they said. In other words, price is used to represent the value of the relationship.

To explore these ideas, the authors designed a model and tested it out in two newly designed conjoint experiments for gift giving. In the first, 99 pairs of friends were asked to each choose from a selection of headphones for themselves, then to choose headphones as a gift for their friend. In the latter case, some participants received no information about their friend’s preference, some were given one piece of information (i.e., one of the choices that the receiver previously made), and some were given full details. This was to manipulate the uncertainty of their decision.

The givers were found to be less price sensitive when buying a gift than buying for themselves, and this was especially the case when they did not have full information about the receiver’s preference. This finding is unique and has not been shown in previous research.

The second experiment looked at the mechanism behind this result – whether price is a signal of relationship importance or a signal of quality. Some 224 pairs of friends were asked to choose from a selection of wireless mice for themselves. Then they were assigned to one of two conditions: select a gift for their friend as in the first experiment, or act as an agent for their friend who needed a wireless mouse but did not have time to go to the shop to buy one.

The results showed that, as with the first experiment, givers relied on their own preferences when they did not have information about their friend’s preferences; price was seen as signaling the value of the relationship; and that signaling value decreased with more information about the receiver’s preferences.

But when participants acted as their friend’s agent, they became more price sensitive, perhaps because they felt the need to be careful with their friend’s money.

The authors decomposed the deadweight loss of their gift-giving findings and found price signaling tended to increase deadweight loss in markets with large variation in prices but relatively small variations in product attributes.

“Interestingly, markets with small differences among alternatives but large price differences are relatively common in the gift industry, for example, jewelry, wines and perfumes. Our results suggest the deadweight loss in such markets is relatively high and that about 77 per cent of this is caused by price signaling.

“This is in contrast to other markets such as books and music, in which products are more differentiated and prices are relatively similar. In these cases, inaccurate predictions of preferences contribute to about 77 per cent of deadweight loss,” they said.

The findings have implications for managers in terms of refining tools for gift recommendations, such as on social media, and investigating whether deadweight loss could be reduced by giving cash or gift cards instead of gifts.