Economic models of consumer behavior almost universally predict that optimal behavior involves aggressive consumption smoothing. That means consumers make themselves as happy as they can be if they find their optimal level of consumption and maintain it month after month, year after year. Fully rational consumers should set consumption today equal to consumption yesterday and so on.
Given this rather robust result, Thanksgiving raises an interesting problem. If you still have a lump in your stomach from dining excess last week, you will immediately recognize it. On Thanksgiving, Americans gather around tables that are positively crushed with the weight of different dishes. They gorge themselves with turkey until their eyes drift shut, and then come back for more. Daily food consumption sees an enormous spike. Our economic models suggest that we would participate in this feeding orgy once, and then find that we could have been happier if we had spread the feast out over many days, that we would experience regrets after such an irrational binge.
Another apparent irrationality exists as well. Whole families gather in kitchens around America and work for hours upon hours preparing the feast. If we value the time spent cooking at the average wage of an American family member, then one can easily impute an implicit cost for all this culinary activity in the thousands of dollars. And yet, we come back the next year and do it again. Why?
This question never occurred to me until this year. Thanksgiving in the Hassett household was a great deal different this year. My mother, the long-time queen of the feast, passed away suddenly last summer, and the management of the festivities fell to the next generation. As the holiday approached, there were conversations in our family that maybe this year we should forego the lengthy preparations and just visit some local restaurant. Why bother sinking all the time and effort into a feast that more than anything may painfully remind us of my mother's absence? Resolution of the question required some careful thinking about the value of the holiday's production.
Looking back over the many previous Thanksgivings, here is what I found.
In most years, our various family members have been scattered far and wide around the country, and have rarely come together during the year except at the two major holidays, Thanksgiving and Christmas. Gatherings outside of the holidays are extremely difficult because other responsibilities get in the way, be they Little League games or business meetings. The holiday serves the function of what economists (following the brilliant work of Thomas Schelling) call a "focal point." We all know ahead of time when the holiday will be, and keep the time clear for a family visit. This lowers significantly the opportunity cost of travel, and raises the benefit of travel as well, since the probability of having a large number of family members gather is higher.
But what about all that cooking? Despite the best efforts of phone conversations, family members are poorly informed about the activities of others through the course of the year. But during the holidays there is ample time for everyone to catch up, and most of the catching up occurs in the kitchen over pie dough and boiling cranberries. While a family cannot sit in a living room and chat for five hours, it is somehow possible to maintain interaction for a similar period while cooking.
From this perspective, then, the entire Thanksgiving makes sense. There is a high fixed cost (travel, etc.) associated with assembling a family. Models that advise consumption smoothing exclude these fixed costs. These fixed costs explain why most families do not smooth the consumption of one another's company over the course of the year. But once the fixed costs have been paid, then it is optimal to maximize our interactions with one another. And the preparation of the feast is part of the consumption. Why, even Homo Economicus himself must have a binge on Thanksgiving.
If one is interested, one could even test empirically this view of Thanksgiving. If Thanksgiving is about family contact more than food then one might expect to see consumers who do not have families smooth their consumption even at Thanksgiving. Those with families will exhibit a sharp spike upward in consumption. That pattern must be in the data.
Without waiting for the evidence, we ate at home this year. Drawing on the many years of experience I acquired watching my mother, I chilled the pastry dough before rolling, roasted the turkey in a brown paper bag, and mashed the potatoes with an ungodly amount of butter. All of that took hours, of course, but the kitchen was crowded with relatives.
Kevin A. Hassett is a resident scholar at AEI.