An Economic and Pedagogical Defense of Gratuities.
A new trend is sweeping the restaurant industry. It isn't the latest India pale ale or avocado toast. Nor is it a novel way to make reservations with a smartphone. Rather, it is a movement to eliminate the reliance on gratuities as a partial form of compensation for waitstaff. Instead, some establishments now rely on a fixed "living wage." Restaurants such as Ivar's Salmon House in Seattle, Bar Marco in Pittsburgh, Chez Panisse in Berkeley, and more than a dozen of Danny Meyer's sit-down restaurants in New York City have jumped on the "no gratuities" bandwagon (Tu 2015; Terenzio 2015; Erbentraut 2015; Passy 2014). Besides the ostensible concern for the waitstaff's financial well-being, other reasons restaurateurs have given for this policy change include customer preference (Erbentraut 2015); avoiding the managerial hassle of dividing gratuities among other workers in the establishment (Passy 2014); happier employees (Terenzio 2015); and even concerns over mitigating sexism and racism (Terenzio 2015). Journalistic accounts of restaurants implementing such a policy usually champion how this new practice has supported the business's financial health, though no systematic empirical study of successes and failures has been conducted to date.
Gratuities aren't just unpleasant for a small set of restaurant owners seeking a better model for compensating waitstaff; they've also been historically unpopular with customers. Cultural historian Kerry Segrave has detailed attempts to eliminate the use of "vails" (tips) to reward servants as far back as the eighteenth century in Scotland and England, though rioting by those servants in London quickly ended that effort (1998, p. 2). Segrave further documents that in "Portland, Maine, in 1905, Mayor Baxter assailed the practice of college students receiving tips as waiters at summer hotels. Public reaction was said to be on the mayor's side" (1998, p. 1). And none other than labor leader Samuel Gompers decried the practice as "blackmail" (Segrave 1998, p. 7). (1) Today, opinion pieces rail against the inefficiency and injustice of gratuities as well (cf. Dunn 2013), and when presented with the option to eliminate tipping from various businesses, most students in my political economy course express support.
But if the no-gratuity or fixed-labor-price model is preferred by a wide swathe of customers, makes employees better off, and enhances a restaurant's profit margin, the obvious question becomes, Why aren't more restaurants adopting this policy? Why does such a seemingly unpopular and inefficient practice as tipping persist? Surely markets would work to eliminate this norm if it was not efficient in an industry as competitive as restaurants. This puzzle presents an excellent opportunity to argue for the efficiency of gratuities, as well as to teach basic economic concepts in an environment that is familiar to most students, whether as restaurant workers themselves, as restaurant patrons, or both.
This essay defends the longstanding practice of tipping in restaurants as an ingenious institutional mechanism for solving three common economic problems: the principal-agent problem; capturing gains from trade via price discrimination; and promoting the cultural trust necessary for anonymous exchange. (2) I argue that the use of gratuities is a win-win-win strategy for restaurant owners, customers, and high-quality employees. Owners benefit by solving the principal-agent problem that bedevils the service industry and by capturing diners who have lower reserve prices for the restaurant experience. Customers win by incentivizing higher-quality service on the margin. And, finally, gratuities act as a separating mechanism that allow high-quality employees to earn more than their low-quality counterparts while signaling to the latter that their talents lie elsewhere. And if poor-quality workers are filtered out of the labor pool, the restaurant industry (or sectors of it) benefits from a higher perception of amenable service. Finally, at the end of this essay, I make the case that teaching students about tipping is a fun and informative method for introducing esoteric economic concepts since nearly everyone in the classroom is familiar with this customary practice. Students who originate from foreign cultures that discourage this norm provide an interesting basis for comparative discussion.
II. The Economic Logic of Tipping
While suffering heavy critique from customers and social critics alike, the institutional persistence of gratuities demands an explanation. Reasons why people leave tips include feelings of fairness and equity, the desire for social approval, and concerns over future service (cf. Conlin, Lynn, and O'Donoghue 2003; Lynn and Grassman 1990). Most of this research focuses on the relationship between the customer and the server, asking why and how much customers tip. Tipping seems to undercut the basic logic of rational cost-benefit exchange in economics, as the custom almost exclusively occurs after a transaction has been agreed upon and carried out. A thinly rational individual who is unconcerned with reputational costs has a strong incentive to skip out, or defect from the norm, without any particular harm (cf. Lynn and Grassman 1990, p. 170). (3) For that reason, psychological explanations for tipping have tended to predominate in the limited literature on the subject. (4) Nonetheless, for three strictly economic reasons--two short term and one long term--tipping is economically rational for business owners, customers, and waitstaff: it solves principal-agent problems; it leverages price discrimination via the gains from trade to capture a broader clientele base; and it builds social trust through a ritualistic signaling mechanism to encourage anonymous trade. The last explanation relies less on "thin" accounts of rationality and dips into the literature on cultural norms, but such an argument can be squared with neoinstitutionalist theory when one accounts for efforts to alleviate uncertainty via the creation of institutionalized behaviors.
A. The Principal-Agent Problem (5)
The boss can't be everywhere. And when bosses are not around, employees have an opportunity and an incentive to shirk. This ubiquitous situation falls under the rubric of a principal-agent problem wherein the interests of an agent (employee) are not fully aligned with those of a principal (boss) who is trusting the agent to perform some duty on their behalf. It's not that the interests of the principal and agent are completely antithetical, but there is discordance between the two. A worker understands that their employment depends on the employer having a successful business and being able to retain and reward the employee, but all things being equal the employee would rather substitute leisure for labor (if paid a salary) or extend the amount of time fulfilling a task (if paid hourly). The trick to solving a principal-agent problem is to design some self-enforcing or policing mechanism that more tightly aligns the interests of the principal and the agent.
The restaurant industry provides an example of the principal-agent problem at play. Diners patronize an eating establishment not only for the food served, but also for the ambience and service. A waiter who mistakes an order, fails to refill beverages, or has a cranky attitude can dampen the pleasure of a night out. But unlike food items, whose quality can be tangibly observed (e.g., proper temperature, appetizing appearance), service is more difficult to assess and is more variable in customer preference. Some diners enjoy attentive service from their waiter so as to catch a show, whereas others prefer to be left alone. A manager or owner could walk the floor to observe whether the waitstaff is performing adequately, but this would not be an efficient use of time.
Enter gratuities. If tipping is a known practice among all diners, the customer becomes the policing mechanism for service quality and rewards it accordingly, freeing the manager to perform other tasks (e.g., ordering more cases of wine). Placing customers in partial control of the waiter's compensation also allows them to signal their desired form of service. To a manager on the restaurant floor, a waiter who is avoiding a table may appear to be giving poor service. However, if the diners at the table signal that they want to be left alone, a waiter who visits the table less frequently to refill water glasses might actually be doing an excellent job. High-quality servers identify patrons' cues and adjust their service accordingly. Customers are just as much the principals as the restaurant manager in this principal-agent scenario. By holding out the possibility of a generous tip at the end of the meal, those patrons can match their desired experience with what the waitstaff delivers. And, of course, repeat customers can signal their desire for a certain type of service by tipping generously with the expectation that their desires will be met in the future. Tipping is a win-win for both the employer and the customer, each of whom acts in the role of principal incentivizing the agent (waiter).
Eliminating tipping in favor of a flat fee for service reduces customers' control over the pricing decision and exacerbates the principal-agent problem. Even if the waitstaff is paid a higher "living wage," which may arguably lead to happier employees and better service, the incentive to customize service for each patron is reduced. If waiters are paid the same amount to tend to different customers, there is less need to read or respond to the mood or preferences of those customers, and customers lose the opportunity to signal via subtle cues or outright communication. Given that checking on whether an order is ready, refilling water glasses, and clearing tables can be a hassle, an employee paid a flat rate will likely reduce this cost by shirking a bit.
This is not to say that good service will come to a screeching halt, as most individuals do take pride in a job well done (cf. Smith  1976), and waitstaff who receive constant complaints will eventually get the manager's attention, but at the margin, we can expect service to deteriorate. A waiter who comes to work with a hangover will put a little less effort into getting the order just right, as they get paid the same whether or not a patron's hamburger is cooked medium or medium well. In fact, curmudgeonly customers or boisterous teenagers may intentionally receive lower-quality service in an effort to ensure that they never return to the restaurant, even against the interest of the owner, who wants the tables filled. After all, who wants to serve fussy clientele? If that fussy client is known to be a good tipper, though, the cost of dealing with the person may be worth the extra effort, and they may grow into a happy customer over time.
The principal-agent problem not only helps us to understand why benefits accrue to the restaurant owner and customers through aligning incentives for good service with the waitstaff, but also shows us where tipping prevails. Gratuities are most effective in an environment where it is difficult to determine service quality, where preferences for style of service are more varied, where it is difficult for a principal to monitor an agent effectively, and/or where customers wish to signal their preferences for future service. Not surprisingly, gratuities are the norm in sit-down restaurants and not in fast food establishments, where it is easy to monitor the quality of service at the counter. We also see tipping used frequently at beauty salons, for car valets, in massage parlors, and for hotel housekeeping service during prolonged stays. Tipping is less common with plumbers, house painters, trash collectors, and university professors. (6) Likewise, a person who frequently orders a customized espresso drink is more likely to tip than somebody paying for self-serve coffee. Perhaps the biggest mystery of tipping remains why someone would tip a taxi driver as a visitor to a large city (where the chances of repeat business with the same driver are nil), or why someone would leave a tip at a roadside diner on a cross-country trip when they know they will never return to that establishment (cf. Azar 2004). I will examine these issues below when accounting for the importance of social norms. The bottom line for tipping and the principal-agent problem is that gratuities are helpful for monitoring and incentivizing high-quality service when the transaction costs of measuring and policing service are high.
B. Gains from Trade and Price Discrimination
Free-market economies run on voluntary exchange when individuals who place differing values on goods and services meet to negotiate the terms of trade. This wondrous, decentralized institution matches resources to their highest-valued uses (Hayek 1945). Granted, some individuals capture more in the gains from trade than their trading partners do, but so long as both parties benefit, a win-win exchange occurs. Nonetheless, each person would prefer to reap more gains from trade than the other partner does.
Enter price discrimination. Price discrimination allows a seller to set a price that most closely matches the buyer's reserve purchase price (i.e., the maximum price at which that person is willing to exchange). Car dealers and real estate agents pepper customers with questions to discover how much they value different types of cars or houses and then direct customers to the product that nets the salesperson a larger share of the gains from trade. Retail manufacturers and stores create variations of similar products to meet the varying reserve prices of different customers. Coupons, happy hours, travel-size containers, and buy two, get one free specials are familiar methods of price discrimination (Landsburg 1989, pp. 314-15). Even the Catholic Church used price-discrimination techniques via private confessionals to determine the differing reserve prices of indulgences for sinners in pre-Reformation days (Ekelund et al. 1996, pp. 161-63).
The trouble with price discrimination, though, is that discovery costs for customers' reserve prices can be high, or the mechanisms imperfect. A coffee shop that wanted to determine how badly somebody needs a jolt of caffeine by haggling with customers would run the risk of slowing down the queue and dissuading customers from purchasing. The competitive nature of markets also mitigates efforts to price discriminate, as rivals can undercut each other's prices. As such, menu pricing (fixed pricing for a good or service) is often the best solution in a world of transaction costs (cf. Coase 1937). Here, the maximizing calculation sets the optimal price that captures the widest customer base at the highest profit rate, thereby optimizing the seller's gains from trade.
Menu pricing runs two risks, though. First, a menu price set too high will exclude potential customers who have reserve prices just below the fixed price, which in turn may lower total revenue, depending on the size of the excluded population. This problem is of special concern to restaurants and taxi services, where assets not in use (e.g., table space, idling autos) represent significant deadweight losses. A seller could reduce the price to his lowest reserve sell price, but he might lose some of the gains from trade if the population with high reserve prices is large. As such, the second risk of menu pricing is setting the price too low and excluding a large number of people who are willing to pay more. In short, setting fixed menu prices is a difficult guessing game. If only there was a way to have customers voluntary sort themselves by their reserve prices.
It turns out, there is! Tipping allows restaurants to price discriminate more effectively on the margin of service, where particular reserve prices may be more variable across different customers and even among the same customers at different times. Everybody wants the most delicious food, (7) but when it comes to service, different customers prefer different types of service and set their reserve prices accordingly. Some customers are very concerned about the service received, whereas others only worry about food quality. And some individuals prefer a very interactive waiter and happily pay extra for chitchat, whereas others demand rapid service so as to be on time to the theater. And these customers may be the same people, but at different times. The problem for the restaurant owner is determining who these individuals are and how to tailor prices to their various desires.
As Hayek (1945) observed, the best knowledge is often local knowledge. Allowing your waitstaff to read the signals of individual customers is an effective way to deliver customized service. The trick is to incentivize the servers to pay attention to such signals and provide service accordingly. With the possibility of a large gratuity for reading the signals correctly and delivering appropriate service, customers gain from being able to send signals and have them read, whereas waitstaff benefit from providing the level of service that matches the customer's willingness to pay (with the added gratuity). With fixed prices and no gratuities, the waitstaff have little incentive to customize service, and the customer's ability to signal a service preference attenuates.
Restaurateurs benefit from tipping, too. Gratuities permit customers to set their own terms of trade for service, allowing owners to reach a more diversified clientele--including both those who place a premium on excellent service and those who are not as concerned with service. As mentioned earlier, filling tables with patrons is a major concern for owners; empty seats represent deadweight loss. If owners benefit by attracting more customers, then waitstaff benefit from more hours and more total income. Figure 1 illustrates how the gains from trade work. The X axis represents prices for the overall dining experience: food, ambience, and service. Ambience is largely fixed, and owners can vary menu items to price discriminate for various food and beverage preferences (e.g., shared-plate pricing) to attract different types of customers. The key challenge becomes how to price discriminate on service.
Customers obviously prefer to pay nothing for the overall experience, but also have a reserve price representing the maximum they will pay. Diners A through E are all different in terms of their reserve prices for the overall dining experience. The restaurant owner maintains a reserve sell price to the left of the chart, which delineates the minimum price at which he is willing to sell the meal. This reserve sell price is set by his material costs (e.g., ingredients, furnishings, building rent) and labor. The overlap between the customers' reserve buy price and the restaurateur's reserve sell price represents the gains from trade. In figure 1, all diners except Person E are willing to patronize the restaurant at the given fixed costs. With a norm of tipping, customers can voluntarily yield some of their gains from trade (up to their reserve buy price) to the waitstaff. (8) Customer A has the widest berth in this regard and could potentially be very generous. Person D has the potential to be the least generous. Nothing here asserts that A will be more generous than D; the potential merely exists. The negotiation of the terms of trade between the server and the customer becomes a fascinating one for economists--a Kirznerian dance through the opaque wilderness of uncertainty with a soundtrack provided by Adam Smith's impartial observer.
Diners may not share all of the gains from trade with the restaurant and waitstaff, but instead may choose to tip to a point where the meal's total cost is below their reserve price. Nothing prevents Customer A from leaving no tip at all, whereas Customer D may share all the gains from trade with the restaurant, making D more generous despite having a lower reserve price. But why would any rational utility maximizer choose to share any of the gains from trade at all? I will return to that question in the next section, but suffice it to say that when the norm of tipping is firmly ingrained within a society, opting out is an empirical rarity. Customers could stiff waiters, and some do, but most do not.
Given that gratuities are the norm, the next question is how much any individual should tip. As a rational customer prefers to keep as much of the gains from trade as possible, there will obviously be downward pressure on the amount left on the table. Yet diners seemingly immune from thin rational calculating may leave large tips (and alcohol certainly helps with generosity). From the restaurant owner's perspective, uncertainty over expected tips makes it difficult to calculate what the waitstaffs base pay should be; from the waitstaff's perspective, high variability in tipping makes a job that relies on gratuities somewhat unattractive. Promulgating a specific "rule" of tipping 15 percent to 20 percent via the usual channels of communicating norms (e.g., parents, peers) helps to alleviate some uncertainty while preserving a customer's flexibility. (9) If the norm surrounding an average tip is 15 percent and is reasonably stable over time and place, the restaurant owner can set wages lower to reflect that anticipated income, keep the overall experience of the meal lower, and perhaps attract customers who may have lower reserve buy prices and are only willing to tip 10 percent. Of course, any waiter would prefer not to serve a low tipper relative to a high tipper, but even a 10 percent tip is better than not having a customer to serve at all.
Figure 2 represents what happens when gratuities are eliminated and owners provide their waitstaff with a higher fixed wage. In this instance, with a fixed menu price for both food and service, a single price point predominates, with no possibility for variation. Customers have fewer degrees of freedom and only can choose to vary their price by ordering lower-priced food options. Depending on how high the increase is for the waitstaff, such a policy runs the risk of moving the restaurateur's reserve sell price above some customers' reserve buy prices. In figure 2, this happens with Customer D, who is priced out of the market, whereas C is right at the margin of dining out or not. One small shock to C's budget (e.g., an unexpected car repair) might shift them below the fixed menu price, resulting in lost business.
While D may not have been the most attractive tipper in the restaurant, D was still a source of revenue for the owner when the norm of gratuities was in place. If enough customers are akin to Person D, the owner could lose enough patrons to force him to reduce workers' hours or lay off waitstaff. A waiter who received a pay raise under fixed pricing but was then laid off or faced reduced hours because of lost business would not benefit here. Furthermore, the restaurateur will want to establish a fixed price that retains as much clientele as possible, which places downward pressure on the new fixed wage rate he sets. Examining figure 2 further, a low fixed price that excludes gratuities will mean that customers with reserve buy prices above the new price do not have to share any additional gains from trade with the waitstaff. This is a potential loss to waiters, represented by the distance between the solid and dashed red lines (with respect to Customer A).
If a significant customer base is willing to share additional gains from trade above the new fixed menu price, the waitstaff will actually be worse off under this new "no gratuities" rule. Moreover, the waitstaff most harmed will be those highly skilled at identifying the preferences of high-reserve-price diners. In other words, the best service providers will be the biggest losers in this fixed-wage system. Poor servers will disproportionately benefit relative to the excellent ones, as their incomes will generally be the same. (10) Managers may be able to differentiate the excellent servers from the poor servers and offer hourly pay raises to the best staff, but this scenario raises the principal-agent problem again.
Managers will have to spend more time on the floor, or surveying customers after a meal, to determine which servers deserve pay raises and which ones need to be fired. The best servers, understanding the difficulty in differentiating the good servers from the bad (a pooling equilibrium), will have an incentive to reduce their marginal effort to the level of the worst server in the restaurant, or to the requirements of the customer who has the lowest reserve price for service (i.e., the customer who doesn't care about service quality as much as food quality). We can expect the restaurant's overall quality of service to decline, which may have the further effect of chasing away "service connoisseurs" who have a higher preference for the entire dining experience relative to just the quality of food. (11) Again, lost business from these customers may necessitate waitstaff being laid off; the policy change ostensibly designed to help employees may have serious unintended consequences.
The restaurants jumping on the "no gratuities" bandwagon all tend to be smaller, locally owned, and/or upscale, catering to a clientele that is less price sensitive (cf. Lynn 2017b). Given the previous discussion, this observation is not surprising. We can assume that the patrons of these establishments are known to have high standards for service and that the employers would have already hired excellent waitstaff worth paying higher wages relative to restaurants with a customer base having more variable reserve buy prices. If the customers' reserve buy prices are known to be high and within a relatively narrow band, there is less risk to the owner of guessing what the new fixed menu price (sans gratuities) should be. It is also easier for the management of boutique restaurants to monitor the quality of their staff, meaning the principal-agent problem is not as salient. This boutique restaurant hypothesis obviously remains an empirical question that can be tested.
All told, the price-discrimination aspect of tipping is a win-win-win for owners, customers, and waitstaff. Owners can maintain lower reserve sell prices and attract a potentially broader clientele while minimizing the deadweight loss of empty tables (cf. Lynn 2017b). Customers can set their own prices according to their own preferences; tipping also makes the dining-out experience more accessible to lower-income individuals. By keeping the menu price of the meal low, and allowing low-income diners to tip at a lower rate (e.g., 5 percent or less), the restaurant can keep the dining experience at a price lower than the diner's maximum reservation price. This further helps to expand the customer base and ensures that tables will not go empty. Diners also do better if the waitstaff have an incentive to guess the customer's reserve price and try to earn the extra shared gains from trade by providing higher-quality service, rather than devolving to standardized service under a "no gratuities" policy. And finally, waiters benefit by ensuring that the restaurant attracts more customers, thereby providing more work opportunities. The most skilled servers will benefit disproportionately by being able to collect gains from trade above what the owner's fixed price for wages would be. Under a system with gratuities, poor servers are likely to be weeded out, as they cannot earn what might be a "living wage." While seemingly a disadvantage for those workers, it is helpful long term to receive signals as to where one's personal talents are best realized. (12)
C. The Moral Economy of Gratuities
In reviewing our discussion on the gains from trade to be shared between customers and service staff, a puzzle arises: Why would any "rational" consumer ever give up more of her gains from trade voluntarily? This seems a particularly mystifying question given that diners leave tips at the end of their experience, after the explicit contracting over food price has already been agreed upon, and it would be easy to stiff the waiter by simply paying the price listed on the receipt. One possible answer is that repeat customers want to signal a desire for future high-quality service, which relates to the principal-agent problem discussed earlier. But what about the eternal puzzle of why customers who will never return to a restaurant still leave tips? As Azar states, "Tipping is consistent with selfish consumers only if they think that tipping today will improve the service they get in the future, since the service today has already been provided. Most people, however, tip even when they never expect to see the tipped worker again, for example when traveling or when tipping taxi drivers" (2004, p. 50, emphasis added). The incentive to not leave a gratuity is inordinately strong for the pure rational maximizer, whom Azar terms "selfish." But consumers are not merely self-interested in ways that give them immediate payoffs. (13) Uncertainty makes the creation of, and adherence to, institutions-including social norms--an economically rational and self-interested solution to larger social problems. If adhering to an implicit social contract makes one better off long term, it is entirely rational and reasonable to abide by societal etiquette (holding constant discount rates).
When I present this puzzle to students in an introductory political economy class, it often elicits answers of the ilk, "We know that waiters are not paid much and we feel sorry for them," an explanation similar to the one proffered by Lynn and Grassman (1990). In response, the challenge is to ask why we don't tip other low-paid workers, such as floor employees at Walmart who help you locate an item, or teens working as ushers at movie theaters who clean up your spilled soft drink. Tip jars at self-serve yogurt shops collect significantly less than a waiter does at a typical family style restaurant. And people are less likely to tip at a coffee shop when they order simple drip coffee as compared to a complex five-adjective specialty drink, even though it is the same barista performing the task.
One would also anticipate that tipping behavior would be associated more closely with those who donate to charities serving lower-income individuals (a hypothesis that has, to my knowledge, not been tested yet). Furthermore, why do we not have a progressive gratuity scale to tip waitstaff at low-end diners (e.g., Denny's) more as compared to the staff at five-star restaurants (e.g., the Four Seasons) who often pull down annual pay above the median income? Azar explains this inconsistency by noting that "the customers of an upscale restaurant are naturally wealthier than those of a family style restaurant. This increases the difference between the income of the tipper and the worker and causes the tipper to feel more empathy and compassion for the worker, encouraging higher tips" (2004, pp. 57-58).
I argue, in distinction to models relying on emotive sympathy, that the answer to this question may lie in a more "thickly rational" moral economy account of our society. First, though, I concede that an explanation relying on sympathy or empathy can be partially explanatory. Adam Smith ( 1976) famously argued that our behavior and moral sentiments are guided by an "impartial observer" who looks over our shoulders. That said, an institutional explanation consistent with economic rationality is equally (if not more) plausible here.
This explanation begins with the simple observation that life is filled with uncertainty, and such uncertainty raises transaction costs, discourages trade, and lowers economic growth. Long term, uncertainty is not good for anybody either in the aggregate or the specific. One way to mitigate uncertainty is to draft explicit contracts anticipating all potential outcomes and provide punishments for contract violations, along with effective enforcement methods. Not surprisingly, that option is inordinately expensive considering the wide range of circumstances--both anticipated and "knowingly unanticipated"--that could arise. A more effective means of managing uncertainty is to provide a general set of norms ("etiquette" or "manners") that guide behavior. As Buchanan notes, "There are countless activities that require persons to adhere to fundamental rules for mutual tolerance, activities that may be observed to go on apace day by day and without formal rules. They go on because participants accept the standards of conduct that are minimally demanded for order to be established and maintained" (1975, pp. 4-5). Order provides predictability and mitigates uncertainty. The argument about adherence to social norms is akin to Buchanan's argument about the implicit constitutional contract within a society. Being born into a nation that has already established a formal set of fundamental laws (i.e., a constitution) does not obligate one naturally to agree to that contract (as you were not party to its initial agreement). Yet we adhere to it, understanding that the basic order it provides is much better than the Hobbesian disorder that would otherwise result (Buchanan 1975).
In market societies, living standards are determined by how extensive the market is, which naturally implies that trade will inevitably occur among "anonymous strangers." Engaging in long-term contracts with individuals whose intentions or trustworthiness you know little about is risky and could deter trade. To mitigate this risk, market societies develop institutional mechanisms such as bonds, insurance markets, and other forms of contract enforcement. Another method of dealing with uncertainty is to create an informal institutional system of ritualistic behaviors that allows individuals to cultivate a reputation for trustworthiness (cf. Greif 1994). Much of early education revolves around inculcating children to such norms, and parents are careful to ensure that their offspring behave well in their local neighborhoods, lest the children's behavior reflect poorly on the parents. Individuals who do not partake in these cultural rituals or normative behaviors provide clues that they should not be trusted in other social interactions (e.g., commercial trade). Such a realization may not be explicit, but rather evolves through a process wherein those who exhibit "proper" manners are rewarded, thereby prompting everybody to imitate them.
Norms function as rules of thumb, allowing humans to make decisions in situations where it is uncertain how to act. They also help to specify property rights in a world where the formal specification of rights is too costly. Norms cannot account for all the possible scenarios we encounter daily, but they can provide "guard rails" that allow for flexibility in making decisions (cf. Bicchieri 2006). They also provide important information about the behavior of others. Observing an individual who shares my etiquette enables me to assume the person shares other qualities similar to mine, thereby reducing my uncertainty about their character and motives. The possibility of deception for personal gain always exists, but the problem is mitigated if one can observe a set of patterned behaviors in realms not directly associated with the specific interaction the two of us are engaged in. If I want to establish a business relationship with a new partner and others tell me that this person is known to be a generous tipper, I know that she is probably willing to give up some ground in the negotiated gains from trade in order to establish a long-term relationship. If, on the other hand, I notice that a potential business partner fails to tip the waiter, it raises suspicions about whether he might defect from an agreed upon contract. (14)
All of this analysis presupposes that norms are a conscious way of solving various informational problems that arise with interpersonal relationships. It does not solve entirely the problem of the utility-maximizing individual stiffing a server at a restaurant they know they will never visit again. But this may simply indicate that norms, once largely inculcated, work best when they are subconscious, helping us to minimize informational costs; we follow a behavioral pattern because we have always done that and need not think about a situation anew at every occasion. Even though the rational calculation to stiff a waiter may be minimal, it is still more effort than merely following the decision rule that we were taught. (15)
Moreover, it is possible that in the backs of our minds we understand that preserving norms of trust and generosity is essential to the functioning of a market system that relies heavily on anonymous transactions. Such an assertion comes close to violating the assumption of methodological individualism that underlies microeconomic theory, but if it is understood that the normative structure of a market economy resembles more of an assurance game than a prisoners' dilemma, then the individual incentives to preserve the cooperative equilibrium make more sense. Again, Nobel laureate James Buchanan (1975) lays this out nicely with respect to why we agree to abide by constitutional rules that we had no part in negotiating for ourselves.
It should come as no surprise that the institution of tipping arose at a time in European history when market economies were expanding (circa the seventeenth century) and originated in a country that epitomized this new capitalist form of economic interaction (i.e., England). Nor should it be surprising, from this "moral economy" framework, that gratuities became the norm in the United States just as the population was becoming increasingly mobile and pluralistic (circa the late nineteenth century) (Segrave 1998). And tipping is not the only cultural norm to be established as a means of mitigating uncertainty. The "cowboy code" that arose in the Wild West during the last three decades of nineteenth century helped to alleviate uncertainty about the intentions of strangers during westward expansion, when it was common for towns to experience a rapid influx of newcomers or to encounter folks who were "just passing through, ma'am" (Adams 1969). Norms associated with this code included tipping your hat to strangers, drinking with your shooting hand, and removing your firearms when sitting down for dinner. (16) This etiquette may seem superfluous and silly to us now, but adhering to such "rituals" sent signals about how you understood and related to other members of the community. We preserve similar normative rituals to this day, including shaking hands when greeting business partners, holding doors open for older individuals, and dressing appropriately at academic conferences. Anyone not adhering to such simple rules probably cannot be trusted in other, more important situations such as commercial transactions involving large sums over extended periods. Tipping is just one of many norms that help people judge an individual's behavior, but also one that solves principal-agent problems and leverages the ability to voluntarily price discriminate.
Overall, a system of gratuities provides a win-win-win for all parties involved, as a society that has strong norms of trustworthiness across the population will have lower transaction costs to commercial activity and, in general, higher levels of economic activity and growth. This need not be the only norm that accomplishes such a task, and there is no particular reason why any society should end up in such a particular equilibrium. Moving toward a "no gratuities" model of restaurant service will not necessarily decrease trust within society, but it does eliminate a longstanding norm.
III. Conclusion: The Benefits of Teaching "Tips"
This essay advanced several theoretical reasons why gratuities are an ingenious cultural and market institution. But the practice of tipping offers more to those within the academic profession. Political economy is filled with numerous concepts that appear odd to undergraduate and graduate students, at least initially. Most expect a course in political economy to be centered around things such as the Federal Reserve, the World Bank, or international trade treaties. Rarely do they come to a course with an understanding that the basic principles of economics are practiced in their daily lives and that these principles are what guide the bigger and more important aspects of life. Being able to instill an appreciation for the economic way of thinking as it applies to their daily lives, and have them evaluate their own behavior, should be one of the primary goals of their introductory education into the field (Heyne, Boettke, and Prychitko 2013; Hall 2014). Using the example of gratuities to provide lessons on principal-agent problems, gains from trade, price discrimination, and social institutions opens a doorway into the realm. Having students think through why they would tip at restaurants also offers some insights into behavioral economics, as students often see their behavior as being emotionally driven. Probing the class as to why they tip in some instances and not in others takes these lessons further and helps students think about the testing of various assertions they make.
Not all students come from nations where tipping is the norm. My experiences with students from various countries in Europe and Asia prove enlightening and further the discussion. For instance, students from China, where tipping is not the norm, often relate how they get sour looks from restaurant employees in the United States when they don't leave a tip during their first visit because they don't know the custom. They further discuss how they need to rely on others to explain the norm, which presents a unique opportunity to open classroom discussion on how norms are disseminated. Several European students note that gratuities are not common in certain parts of Europe, such as the Nordic countries, but then reveal that they often leave a little bit of extra money on the table at restaurants that they visit regularly, emphasizing the strategic nature of tipping. These discussions present joyful "a ha!" moments in the classroom when students realize that their everyday behaviors conform to economic principles they thought they were immune from.
As noted previously, tipping is only one of a wide array of potential institutions that can build trust within a society. The stories of how foreign students adapt to new cultural norms help inform us about how culture is malleable and ever changing. Often, these students' anecdotes illustrate indirectly many of the points made earlier or push the conversation into new areas that open students' minds to the broad applicability of economic reasoning. I have had students from Nordic countries note that gratuities are mandatory (which is roughly equivalent to a no-gratuity policy that builds in a higher wage), and yet regulars to local establishments will tip voluntarily on top of the mandated gratuity "to insure promptitude" in the future. (17)
The recent wave of no-gratuity establishments makes this topic all the more relevant, as many students rally around such a movement for "social justice" reasons, or just to make the dining experience less stressful. Questioning students as to why all establishments haven't adopted (as of yet) this no-gratuity model allows for comparative analysis, prompting us to think about how and why institutions change over time. This essay made the theoretical case that tipping is a socially efficient institution that benefits employers, customers, and the employees who receive gratuities. Not only can tipping benefit the highly skilled waitstaff in a restaurant, but it can signal to those who are not as skilled at waitering that they should deploy their talents elsewhere, a lesson in how markets send signals about value. Of course, these theoretical arguments may be prove to be empirically shaky when put to the test, and students may reject them at first glance based on how they want to perceive their own behavior, but such challenges only make discussion more lively and may suggest a variety of interesting student research projects that could be conducted. Thus, the social institution of gratuities is not only a win-win-win situation for employers, customers, and employees, but a win-win-win-win-win when one includes professors and students!
Adams, Ramon. 1969. The Cowman and His Code of Ethics. San Antonio: Encino Press.
Alchian, Armen. 1950. "Uncertainty, Evolution, and Economic Theory." Journal of Political Economy, 58(3): 211-21.
Azar, Ofer H. 2003. "The Implications of Tipping for Economics and Management." International Journal of Social Economics, 30(10): 1084-94.
Azar, Ofer H. 2004. "What Sustains Social Norms and How They Evolve? The Case of Tipping." Journal of Economic Behavior and Organisation, 54(1): 49-64.
Bicchieri, Cristina. 2006. The Grammar of Society: The Nature and Dynamics of Social Norms. Cambridge: Cambridge University Press.
Boettke, Peter, W. Zachary Caceres, and Adam Martin. 2013. "Error Is Obvious, Coordination Is the Puzzle." In Hayek and. Behavioral Economics, ed. R. Frantz and R. Leeson, 90-110. New York: Palgrave.
Buchanan, James. 1975. The Limits of Liberty: Between Anarchy and Leviathan. Chicago: University of Chicago Press.
Coase, Ronald. 1937. "The Nature of the Firm." Economica, 4(16): 386-405.
Conlin, Michael, Michael Lynn, and Ted O'Donoghue. 2003. "The Norm of Restaurant Tipping." Journal of Economic Behavior and Organisation, 52(3): 297-321.
Dunn, Elizabeth Gunnison. 2013. "Why Tipping Should Be Outlawed." Esquire, June 19.
Erbentraut, Joseph. 2015. "Here's What Really Happens When a Restaurant Bans Tipping." Huffington Post, October 21.
Ekelund, Robert B., Robert F. Hebert, Robert D. Tollison, Gary M. Anderson, and Audrey B. Davidson. 1996. Sacred Trust: The Medieval Church as an Economic Firm. Oxford: Oxford University Press.
Filloon, Whitney. 2016. "Restaurants Can't Make Servers Share Tips with Kitchen Staff, Court Rules." Eater.
Greif, Avner. 1994. "Cultural Beliefs and the Organization of Society: A Historical and Theoretical Reflection on Collectivist and Individualist Societies." Journal of Political Economy, 102(5): 912-50.
Hall, Joshua, ed. 2014. Homer Economicus: The Simpsons and Economics. Stanford, CA: Stanford University Press.
Hayek, F. A. 1945. "The Use of Knowledge in Society." American Economic Review, 35(4): 519-30.
Heyne, Paul, Peter Boettke, and David Prychitko. 2013. The Economic Way of Thinking, 13th ed. New York: Pearson.
Hirschman, Albert O. 1970. Exit, Voice, and Loyalty: Responses to Decline in Firms, Organisations, and States. Cambridge, MA: Harvard University Press.
Landsburg, Steven E. 1989. Price Theory and Applications. Orlando: Dryden Press.
Lott, John R., and Russell D. Roberts. 1991. "A Guide to the Pitfalls of Identifying Price Discrimination." Economic Inquiry, 29(1): 14-23.
Lynn, Michael. 2017a. "Does Tipping Help to Attract and Retain Better Service Workers?" Journal of Foodservice Business Research, 20(1): 82-89.
Lynn, Michael. 2017b. "Should US Restaurants Abandon Tipping? A Review of the Issues and Evidence." PsychosociologicalIssues in Human Resource Management, 5(1): 120-59.
Lynn, Michael, and Andrea Grassman. 1990. "Restaurant Tipping: An Examination of Three 'Rational' Explanations." Journal of Economic Psychology, 11(2): 169-81.
Lynn, Michael, J. Robert Kwortnik, and Michael C. Sturman. 2011. "Voluntary Tipping and the Selective Attraction and Retention of Service Workers in the USA: An Application of the ASA Model." International Journal of Human Resource Management, 22(9): 1887-1901.
McKenzie, Richard B. 2008. Why Popcorn Costs So Much at the Movies: And Other Pricing Pupples. New York: Copernicus Books.
Passy, Charles. 2014. "The End of Restaurant Tipping?" MarketWatch, March 12.
Segrave, Kerry. 1998. Tipping: An American Social History of Gratuities. Jefferson, NC: McFarland.
Sisk, David E., and Edward C. Gallick. 1985. "Tips and Commissions: A Study in Economic Contracting." Working paper no. 125. Washington, DC: Bureau of Economics, Federal Trade Commission.
Smith, Adam. (1759) 1976. A Theory of Moral Sentiments. Indianapolis: Liberty Fund. Terenzio, Olivia. 2015. "Why the Country's Top Restaurants Are Eliminating Tips." OpenTable, October 23.
Tu, Janet I. 2015. "Ivar's to Raise Restaurant Workers' Wages to $15 Right Away." Seattle Times, March 30.
Political Science University of Washington
(1) Segrave attributes this statement to Gompers's 1910 book, Labor in Europe and America, but does not cite a page number.
(2) The origins of tipping are murky but likely date back to the late medieval period in Europe, particularly England, and then to American colonists in the 1600s, and then to the US Civil War, when the practice gained prominence (Segrave 1998).
(3) Azar (2004), Lynn and Grassman (1990), and Conlin, Lynn, and O'Donoghue (2003) argue that survey evidence (from a limited set of studies) does not show that people tip with the hope of receiving better service in the future, though questionnaires for these studies are conducted face-to-face with restaurant patrons directly after they finish a meal. In such surveys, it is reasonable to assume that the socially acceptable answer is that one is concerned about the server's personal welfare and income, and not a more self-serving answer that they are paying for future service. Moreover, these studies do not directly test whether a customer fears reputational damage if they don't tip, which is slightly different than trying to purchase future service. These findings seem counterintuitive in that I have talked casually with many bartenders and waiters who have assured me that customers who are well-known, generous tippers receive more customized attention than those who tip poorly. I have casually talked with customers who understand this concept as well. Armen Alchian (1950) has pointed out that in a world of uncertainty, we often do not understand explicitly the mechanisms that allow us to engage in maximizing behavior; rather, more efficient institutions tend to persist over time and become adopted "unthinkingly." As such, straightforward questions asking respondents why they engage in a particular behavior may not reveal the underlying rational economic mechanism at work, and instead may merely provide a socially acceptable justification of a norm taught to us by our parents and peers. I am indebted to an anonymous reviewer for this insight and reference.
(4) Michael Lynn, a professor at Cornell University's School of Hotel Management, has written the bulk of academic research articles on gratuities, mostly in restaurant settings. His work is conversant with a basic understanding of microeconomics by way of the behavioral economics literature. Azar (2003) also operates from a behavioral economics approach to this question and draws similar conclusions.
(5) It is surprising to note that the principal-agent problem arising between a restaurant owner/manager and the waitstaff is considered rarely in explanations for why tipping occurs; the literature seems to focus exclusively on the customer-server nexus of the relationship. A few exceptions exist (cf. Sisk and Gallick 1985), but these focus on whether higher-skilled employees remain in the service sector relative to those with poorer service-based skills and not on the motivations employers have for solving the principal-agent problem (cf. Lynn 2017a; Lynn, Kwortnik, and Sturman 2011).
(6) An anonymous reviewer of this article noted that in some areas where there is private trash pickup, some households will leave a case of beer for the sanitation workers. I thank that reviewer for bringing this practice to my attention and will adjust my future interactions with Waste Management accordingly. And professors do sometimes get small gifts from students, most of whom have been in several classes and/or need letters of recommendation.
(7) Restaurants, of course, find creative ways to price discriminate on the dimension of food and beverage products. For example, wine and desserts are often priced to provide a higher profit margin than entrees, as customers who order those products often linger longer, taking up valuable table space. This pricing strategy may not always pay off, particularly if the cost of the valuable table space exceeds the added profit margin on the dessert or wine, but as McKenzie notes, "It is very tough for many restaurants ... to identify the price sensitivity of individual customers (say, by their looks or dress) as they walk through the doors. This doesn't mean that they can't price discriminate; they only have to develop a less ambitious strategy than charging every customer a different price for each unit bought" (2008, p. 64). He further notes that "some of the price differences for various sizes can be chalked up to differences in the cost of packaging (a widely recognized explanation) ... some of it can also be chalked up to the fact that stores are 'walking their customers down their demand curve (a not-so-appreciated explanation)" (2008, p. 62). While the latter sentence is in reference to different-sized packages of beans and candy, it is just a small step to realize that restaurants and other services (e.g., airlines) are also selling "packaged" goods. An anonymous reviewer pointed out that Lott and Roberts (1991) tend to see less price discrimination in such variable prices and explain the differences due to often subtle and unseen production costs. I favor the view proffered by McKenzie that there is likely a mix of both.
(8) I am aware that some restaurants choose to have gratuities shared between waitstaff and the kitchen, although such "tip pooling" is illegal in several US states (Filloon 2016). I am keeping the model relatively simple, but generalizing the model for how tips are distributed is a fairly easy task. A discussion of why tip-sharing may or may not reduce the quality of service would make for an excellent future research project and classroom exercise.
(9) A number of interesting norms surround tipping. At the many taverns I have inhabited, it is typical that the minimum tip you leave for any drink is one dollar, despite the drink being perhaps $3. I also learned of a subtle signaling mechanism wherein a customer will leave a penny along with the dollar tip, as a dollar tip on a $3 drink (a 33 percent gratuity) could be interpreted as the customer being very happy. If the penny is placed face up, it signals that the service was good, but if it is tails up, the service was bad. Why leave a tip at all if one isn't satisfied with the service? One doesn't want to be a jerk in front of friends or, if they ever intend on returning, to the bartender. The tails-up penny is designed as a first warning about service.
(10) Lynn (2017a) argues that tips do not help in retaining high-skilled workers in a study that relies on a nonrandom (snowball) survey of roughly 694 food servers and their self-reported attitudes towards gratuities. Data collection was as follows: "Several different writers of restaurant server blogs were asked, and agreed, to post a link to the survey and to encourage their readers employed as waiters/waitresses to complete it. In addition, the end of the survey asked respondents to encourage other servers they knew to complete the survey" (Lynn 2017a, p. 83). While potentially interesting information, Lynn does not track actual employment data. It is possible that high-skilled servers moved to more exclusive restaurants that valued their talents more. If that were true, gratuities would still operate as an incentive to retain high-skilled servers within the entire restaurant industry (as compared to being retained in one particular establishment).
(11) The connoisseurs of service may also be the individuals who are first to complain about the drop in quality before exiting, depending on the opportunities to exit to higher-quality restaurants. See Hirschman (1970, pp. 138-51) for a discussion on the role of connoisseurs in the exit-voice dynamic.
(12) One other thing worth mentioning is that under current regulatory policy in many states, it is cumbersome to fire a worker, lest claims of discrimination or other such accusations arise. A system of gratuities allows individual staff members to get feedback on their own quality irrespective of what their boss says, and then decide if it is worth it to pursue continued opportunities in this line of work. Waiters who cannot earn sufficient tips are likely to become discouraged and "fire themselves," saving the employer a great deal of grief.
(13) Behavioral economists are quick to critique rationalist accounts of human behavior on the observation that human beings are not perfect calculators. While many models of microeconomic behavior use such constraining assumptions, there is also a wide recognition that the ubiquitous reality of uncertainty in life limits an individual's calculating power. Whereas behavioralists tend to see psychological solutions to informational limits, rationalist accounts include the ability of humans to realize their cognitive limitations and design institutions that overcome some of these problems (Boettke, Caceres, and Martin 2013).
(14) Tipping has a strong signaling component to it when interacting with other individuals. On a first date, and in less gender-enlightened times, the male would traditionally leave a generous tip and make that known to his female partner. This behavior indicated that he was willing to sacrifice and share his wealth with others, a sign of a good partnership in the future.
(15) The infamous scene with Steve Buscemi in Reservoir Dogs comes to mind here. Buscemi's character lays out the case for the thinly rational calculation regarding gratuities, but his compatriots stand in disbelief because that is not how things are to be done. There is even normative honor among the most vile of thieves, indicating the power that "codes of conduct" have over the "ordered anarchy" of our daily lives (cf. Buchanan 1975, p. 20).
(16) Dinner tables seem to be an agreed upon "safe space" (to use modern parlance) where individuals who may be in conflict with one another can negotiate settlements peacefully. Removing weaponry indicates a willingness to do this. I observe similar patterns with the mafia as depicted in movies; if anyone violates the norm of removing weapons at the table and engages in violence, the result is usually an all-out gang war, a situation that neither party usually wants.
(17) For what it's worth, it is unclear whether the word "tip" initially was an acronym for "to insure promptitude," as the etymology is murky. Segrave (1998, p. 4) notes that British author Samuel Johnson saw those words printed on a "tip jar" at a restaurant in the mid-eighteenth century.
Caption: Figure 1 Gains from Trade with Gratuities
Caption: Figure 2 Gains from Trade without Gratuities
|Printer friendly Cite/link Email Feedback|
|Publication:||Journal of Private Enterprise|
|Date:||Feb 8, 2018|
|Previous Article:||Removing the 1970s Crude Oil Price Controls: Lessons for Free-Market Reform.|
|Next Article:||Austrian Macroeconomics in Search of Its Uniqueness.|