*Some Limitations of the Happy-Productive Worker Thesis*

The ambiguity in the studies on the HPWT can be explained in part by the limitations of these studies [26]. First, they focus on hedonic constructs of wellbeing at the expense of eudaimonic wellbeing. In fact, most of the research has studied wellbeing from the hedonic perspective, understanding it as global evaluations of satisfaction (e.g., job satisfaction). More recently, valuable studies have revisited the thesis of the happy and productive worker, studying the possibility of expanding it conceptually to include affect [7] or alternative relationships between satisfaction and performance [8] by evaluating affective wellbeing, both as a state and a trait [19]. However, this thesis has not been extended to consider key wellbeing constructs, such as its eudaimonic dimension, which involves purpose and personal growth. Wellbeing has also been conceptualized as an eudaimonic experience of meaning at work and purpose in life [27]. This conceptualization of subjective wellbeing can be reflected in the recent progress in its measures [28], which distinguish between activities that people consider 'pleasurable' as opposed to the 'worthwhileness' or meaning at work associated with these activities [2,29,30]. Although few studies have investigated the relationship between eudaimonic wellbeing and performance [31], some research suggests that this relationship exists. For example, Niessen et al. [32] demonstrated that, on days when employees had increased perceived meaning at work, they reported being more focused on tasks and behaving in a more exploratory way, compared to days when they evaluated their work as less meaningful to them.

A second limitation is that, in the study of the relationship between happiness and productivity, little attention is paid to a precise operationalization of productivity, and even its operationalization as job performance is far from systemic and comprehensive in terms of its dimensions or facets (e.g., in-role performance, extra-role performance, creative performance). Job performance can be understood as "a function of a person's behavior and the extent to which that behavior helps an organization to reach its goals" [33] (p. 187). However, there is considerable debate about what work performance is. Koopmans and colleagues [34], in their systematic review, observe that, according to different studies on work performance, it can be conceptualized using the following broad dimensions: task performance, contextual performance, and counter-productive behavior. Task or in-role performance is intrinsically related to the activities included in the job description. Contextual performance refers to behaviors that are not directly related to the activities included in the job description. Organizational citizenship overlaps with the definitions of contextual performance and refers to helping others at work in the social and psychological context, thus promoting task

performance [35]. Counterproductive work behaviors include behaviors such as absenteeism, theft, and substance abuse. Furthermore, creativity [36] and innovation [37] have been pointed out as another important aspect of job performance. Several authors suggest conceptualizing job performance using a broader theoretical framework, in order to mitigate error sources and find relationships between performance and job satisfaction [38]. In the present study, we incorporate different aspects of performance (in-role, organizational citizenship, and creative performance) in a global measure. Performance evaluations may come from different sources (e.g., self-assessed, supervisor, peers, customers, etc.). It is necessary to complement the employees' self-rated performance assessment with the supervisors' evaluation of their performance in order to avoid employee leniency or self-deception in self-ratings, which has been shown to be particularly prominent in overall or general performance assessments [39]. By including supervisors' evaluations of their employees' performance levels, we make sure that we are using evaluations that have been shown to have the highest mean reliability, as found in a meta-analysis by Conway and Huffcutt [40]. Therefore, the present study, in addition to employees' self-ratings of their own performance, includes information about their performance from their direct supervisors.

A third limitation lies in the fact that most organizational research has studied "happiness" as an antecedent of "productivity", and only a few studies have looked for the inverse relationship [24,31]. However, there is evidence suggesting that work performance can explain wellbeing indicators. For example, evidence shows that self-rated performance predicts an increase in dedication and a decrease in emotional exhaustion over time [41]. Moreover, performance [42,43] and the experience of making progress toward one's goals at work [44–46] have been shown to predict positive affective states. Additionally, studies have shown that personal initiative is positively related to an increase in work engagement over time [47]. Along the same lines, there is evidence that on days when employees were strongly focused on tasks at work, they also exhibited more vitality and learning than on days when they were weakly focused on their tasks [32].

A fourth limitation is that the studies from both the happy-productive and productive-happy approaches have assumed positive linear relationships, although other patterns of relations may exist, especially those that establish negative relationships between these two variables. These complex and alternative relations between these constructs require taking into consideration different configurations or patterns of these relationships, instead of analyzing them sequentially. In fact, the studies carried out within the happy productive thesis emphasize the results that confirm this thesis. These studies tend to especially explore the synergistic side of the model that produces a win-win situation for employers and employees (happy and productive), while disregarding the antagonistic or win-lose relations (happy and unproductive or unhappy and productive). However, some studies suggest that we should pay more attention to these antagonistic relations, showing, for instance, that difficulty in remembering information and poor task performance can be considered negative consequences of being "happy" at work [48]. Furthermore, other authors provide evidence of the benefits of negative affect on creative performance [49]. Based on this research, Peiró et al. [26] proposed the need to attend to not only the synergetic relations between performance and wellbeing, but also to the antagonistic ones, thus extending the propositions of the HWPW. They proposed the coexistence of four patterns of relationships between performance and wellbeing: "happy-productive", "unhappy-unproductive", "happy-unproductive", and "unhappy-productive". In fact, Ayala et al. [50] found support for these different types of patterns when considering job satisfaction and innovative performance in young employees. Moreover, they found that almost 15% of a sample of Spanish young employees fell in the group of unhappy-productive (about 9%) or the group of happy-unproductive (more than 5%). Acknowledging that the correlations between happiness and productivity are moderated, it is important to focus on the different groups of workers according to their profiles. In order "to learn more about individuals who are outside the hypothesized pattern ... , it is now desirable to investigate additional measures of wellbeing and performance and identify situational and personal features associated with membership in each cluster" [51] (p. 12).

In order to overcome the limitations of the research mentioned above, in the present study, we address them by revisiting the happy-productive worker, incorporating both the hedonic and eudaimonic components of wellbeing and considering different aspects of job performance as well as different evaluation sources. In addition, in this study, we consider wellbeing and performance simultaneously, instead of analyzing the sequence between these two constructs. To this end, we study patterns of wellbeing and performance that serve to integrate these two constructs, taking into account different operationalizations where neither of them is an antecedent of the other, in order to identify different patterns of employees, both synergistic (i.e., happy-productive) and antagonistic (i.e., unhappy-unproductive, happy-unproductive and unhappy-productive). In this way, we aim to further advance our knowledge in the direction pointed out by Warr and Nielsen [51] when they proposed identifying situational and personal features associated with membership in each cluster. More specifically, we formulate the following research questions:

