Linguistically Induced Time Perception and Asymmetric Cost behavior.
| Verfasser | Huang, Wei |
| Posten | RESEARCH ARTICLE |
1 Introduction
Prior studies in cost behavior show that costs decrease to a lesser extent in response to sales decreases than they increase in response to an equivalent level of sales increases. This asymmetric cost behavior (i.e., cost stickiness) occurs because managers make deliberate decisions to commit resources, and changes in resource commitments incur resource adjustment costs. When sales decrease, managers may choose to retain some unused resources rather than incur resource adjustment costs, such as severance payments to dismissed employees and disposal costs for capital equipment. When sales increase, managers have less room for discretion and must add needed resources (Anderson et al. 2003).
A conceptual innovation that opens up the "black box" of cost behavior is to view managerial discretion as the fundamental driver of the costs. Such a decision-based approach allows researchers to examine how managers' beliefs and psychological biases affect cost behavior. A number of studies examine behavioral factors that determine the degree of asymmetric cost behavior, including managers' optimism or pessimism (Anderson et al. 2003), managerial overconfidence (Chen et al. 2018), and management expectations measured by the tone of forward-looking statements (Chen et al. 2019). In this paper, we examine whether linguistically induced time perception relates to the degree of asymmetric cost behavior. Specifically, we examine whether grammatical differences in how languages separate future events from present ones are associated with asymmetric cost behavior in different countries. Given that human beings are exposed to languages earlier than they are exposed to any other potential factors, such as religion, law, regulation, and organization, we expect languages to have a profound effect on managerial behavior and corporate outcomes.
Languages differ in how they encode time. Speakers of strong time reference (TR) languages (e.g., English) mark present and future events distinctly, while speakers of weak-TR languages (e.g., German) do not. Economics research finds that differences in language regarding how speakers encode time can affect people's economic behavior. Chen (2013), for example, shows that speakers of weak-TR languages exhibit more future-oriented behavior, namely saving more, exercising more, and smoking less, than speakers of strong-TR languages.
Building on dynamic factor demand models developed in labor economics, we posit that TR in languages affects asymmetric cost behavior. We illustrate a managerial resource commitment decision as a multi-period decision model for an input factor. In this model, managers make an optimal factor choice or adjustment when sales change. When sales increase, managers must add resources to meet the demand and avoid opportunity costs arising from resource shortfalls in the current period. In this case, the role of time perception is not significant. When sales decrease, however, managers have more discretion, as the tradeoff will be influenced by managers' expectation about the permanence of revenue declines. We show that linguistically induced time perception affects this tradeoff because, while it does not affect speakers' evaluation of current costs, it influences their evaluation of future costs.
Prior studies (e.g., Chen 2013) suggest that, compared to speakers of strong-TR languages, speakers of weak-TR languages apply lower discount rates. As they apply lower discount rates, weak-TR-language speakers will perceive the present value of future costs to be greater. In the event of a sales decrease, lower discount rates lead managers to perceive a greater degree of both (1) future upward resource readjustment costs or potential opportunity costs (costs that arise when demand picks up again), and (2) future downward adjustment costs (adjustment costs that arise when demand does not pick up again). However, a lower discount rate will have a greater impact on the former than the latter to the extent that the former is larger than the latter.
When demand rebounds, managers have to quickly acquire resources even if it means that they must pay a premium, such as overtime pay and/or express fees, to restore capacity. Otherwise, capacity shortfalls lead to high opportunity costs arising from the inability to capture high demand. Therefore, managers' tendency to retain unused resources will be greater when she/he applies a lower discount rate to future payoffs. Thus, compared to strong-TR-language speakers, weak-TR-language speakers are more likely to hold on to slack resources when sales decrease, leading to a greater degree of cost stickiness in countries with weak TR in languages.
Using a large number of firm-year observations from 41 countries between 2002 and 2014, we find that weak-TR languages are positively associated with cost stickiness. That is, cost asymmetry is more prevalent in countries with weak TR in languages than in countries with strong TR in languages. We conduct extensive robustness checks in the context of existing literature, such as using different types of costs, controlling for earnings management or labor market regulations, employing alternative model specifications, replacing a dichotomous classification of TR in languages with continuous measures of TR based on a word-frequency analysis of online texts, controlling for an alternative measure of long-term orientation or alternative measure of managers' optimism. We find that our results remain robust to above sensitivity analyses.
Our study contributes to the literature on three fronts. First, we add to the literature on asymmetric cost behavior by documenting the systematic relation between linguistically induced time perception and asymmetric cost behavior. Second, our study also contributes to the literature on the cross-country differences in asymmetric cost behavior. We show that the grammatical structure of languages is associated with cross-country differences in asymmetric cost behavior. (1) Finally, our study adds to the emerging literature that studies the effects of languages on corporate actions and managerial choices. We provide detailed discussions in conclusion.
2 Literature Review and Hypothesis Development
2.1 Linguistic Relativity Principle
The linguistic relativity principle, or the Sapir-Whorf hypothesis (Whorf 1956), holds that languages can influence human cognition, so that individuals' thoughts and behaviors are affected by the languages they use. This principle posits that the structures of different languages influence how their speakers perceive and conceptualize the world. Although Chomsky (1957) challenges this idea and suggests that human beings have innate and genetically determined language abilities, abundant evidence supports the Sapir-Whorf hypothesis. Several studies, for example, show an association between languages and color cognition. (2) Brown and Lenneberg (1954) report that Zuni, a language of the Zuni people, does not have a lexical distinction between yellow and orange, and that its speakers have difficulty recognizing nuanced differences between yellow and orange. Winawer et al. (2007) report that Russian speakers discriminate dark blue from light blue better than English speakers because the Russian language distinguishes obligatorily between these two shades of color.
Languages differ not only in color terms but also in whether they require speakers to mark future events grammatically. Germanic languages (except for English) do not require speakers to mark future events grammatically when making predictions. A German speaker forecasting a freeze, for example, can do so in the present tense, as in: "Morgen ist es kalt" which translates to "It is cold tomorrow." In German, it is common to describe future events without future TR. English, in contrast, requires the use of a clear future marker, as in: "It will be cold tomorrow" (Chen 2013). (3) That is, English requires speakers to encode future events with future TR, while German does not.
Speakers of languages without obligatory future marking narrate future events in the present tense, which leads weak-TR-language speakers to perceive future events to be less distant. Chen (2013) provides evidence that weak-TR-languages speakers exhibit more future-oriented behavior than strong-TR-language speakers. In particular, Chen (2013) finds that weak-TR-language speakers save more, exercise more, and smoke less than strong-TR-language speakers. Sutter et al. (2015) also document the positive relation between weak-TR languages and future-oriented behavior. In an intertemporal choice experiment with elementary school children in an Italian city where one half of the population speaks Italian (strong-TR language), and the other half speaks German (weak-TR language), Sutter et al. (2015) find that German-speaking children are more likely to delay gratification than Italian-speaking children.
2.2 Asymmetric Cost Behavior
In the not-so-distant past, researchers assumed that a change in costs is proportional to a change in activity. Studies in recent decades, however, show that cost behavior deviates from the traditional model. Cooper and Kaplan (1992) and Noreen and Soderstrom (1997) document that it is harder to lower costs in times of a fall in activity level than to increase costs in times of a rise in activity level. Anderson et al. (2003) find that the extent of a decrease in costs in times of a drop in sales is smaller than the extent of an increase in costs in times of an upturn in sales of the same proportion.
Anderson et al. (2003) posit that asymmetric cost behavior occurs because changes in resource commitments incur resource adjustment costs, and managers make deliberate decisions to commit resources. Adjusting resources in the short run incurs substantial resource adjustment costs such as severance payments to dismissed workers and disposal costs for capital equipment. When sales...
Um weiterzulesen
FORDERN SIE IHR PROBEABO ANCOPYRIGHT GALE, Cengage Learning. All rights reserved.
